How common are gastrointestinal disorders in children with autism? Kuddo T and Nelson K B Question: Are children with autistic disorders more likely to have stomach and intestinal disorders than children who do not have an autistic disorder? Background: In the medical literature, the print press, on Web sites, and in Internet chat rooms, some research and parental reports of stomach and intestinal symptoms in children have fed a belief that children with autism are more likely to suffer from gastrointestinal disease than are typically developing children. The review examined the existing research literature on the subject to see if there is solid evidence to support this view. Data Sources: The authors searched PubMed/MEDLINE from 1966 to December 31 2002, using the terms "autism" or "autistic", and "gastrointestinal" or "bowel" with an age limit of 0 to 18 years. They also searched the reference lists of the articles retrieved and then looked at the abstracts from the 2000 Congress of Pediatric Gastroenterology, Hepatology, and Nutrition poster session to find as yet unpublished data. Study Selection: Process is not described. Data Extraction: The authors noted the study design, source population, numbers of affected vs. unaffected children, whether the comparison groups were age and sex matched, whether they were matched on I.Q., and they then noted the prevalence of GI symptoms. Main Results: Of 91 articles identified, 3 population studies were examined, as were two referral population studies. It was difficult to compare the studies or combine their results because most lacked clear definition of GI symptoms and they did not always note the length of time children were having GI symptoms. In the population studies, Black (2002) found the rates of GI symptoms were 9% vs 9% in controls; Taylor (2002) found a rate of 17% in a study that did not use controls, and Fombonne (2001) found a rate of 18.8% in a study that did not use controls. In the referral populations, using surveys, Horvath et al (2002) found rates of 80.2 to 84.1% vs controls with a rate of 31.2% and Melmed (2000) found a rate of 46%, with no controls. Conclusions: There is no solid research evidence to support the theory that children with autism have a greater likelihood of having GI symptoms compared to typically developing children. The studies thus far have not been done with representative groups of children with autism and the studies lacked appropriate controls against whom to compare rates of GI symptoms. In instances where the researchers used case-control methodology, the healthy controls would have to have been carefully matched and drawn from the same population pool as the children with autism. As well, the control children with GI symptoms but without autism would have to have similar kinds of symptoms and the same range of symptom severity as the children with autism who also had GI symptoms. It is important to note that data from gastroenterologists' practices are biased because their patients are a special group who may not share characteristics of children whose GI symptoms are not severe enough to require referral, or they may differ from children who live in geographical areas without access to specialists.
The authors cite drug and dietary effects that can produce abdominal pain, diarrhea, and constipation. They also state that constipation and diarrhea are very common in all toddlers. In summary, there are currently no good studies that demonstrate that there is an increased risk of having gastrointestinal disease.
Current Opinion in Pediatrics 2003, 15: 339-343.