As many readers may know, one of the ideas about autism promulgated by Andrew Wakefield and other anti-vaccine proponents is that autistic behavior is a response to gastrointestinal problems, and that those problems may be caused by the actions of vaccines in the digestive tract. The American Academy of Pediatrics recently published a statement about the gastrointestinal issue, and various comments have been published. It's interesting to see what conclusions have been drawn about this by different people. Let's look at a few comments on this topic.
At the "Age of Autism" web site (www.ageofautism.com/2010/01/national-autism-association-on-pediat...), Lori McIlwain, the National Autism Association board chair, is quoted as making the following statement with respect to the AAP statement: "We are finally getting mainstream acknowledgement that our kids are physically sick, and not the victims of some mysterious genetic behavioral disorder." Other comments on the same page suggest that the AAP report is in agreement with parents who have believed that medical treatment of gastrointestinal disorders would be an effective way to improve the condition of autistic children.
Is this what the AAP report actually says? In fact, the published article, by Buie and many other authors (Evaluation, diagnosis, and treatment of gastrointestinal disorders in individuals with ASDs: A consensus report. Pediatrics, 2010, Vol. 125 (Supplement), pp. S1-S18) stresses the point that relevant, evidence-based recommendations are not yet available, but that autistic children, like all children, should have thorough evaluation and treatment for physical symptoms. Buie and his co-authors note that autistic children can have the same gastrointestinal problems as the ones that are common in the rest of the population, but that there is no clear evidence for the idea that autistic individuals have a special type of gastrointestinal disturbance that is not found in other people. They also point out that it is possible for pain or discomfort to set off "problem" behaviors, which might make more sense to us if the sufferer were able to communicate better than most autistic children do.
Basically, the AAP report asks for care in dealing with possible medical problems of autistic children, and serves as a reminder that behavioral treatment can be an incomplete response to the needs of people for whom behavioral problems are only one part of their difficulties. This is an important statement, especially for children whose treatment is largely in the hands of practitioners with backgrounds in education, psychology, or social work, fields that do not necessarily give much consideration to physical health. The report also emphasizes the need for appropriate diet to support growth and development, and the difficulty in providing this when children have the very narrow food preferences that are not unusual among autistic individuals; Buie and his co-authors point to the appropriateness of following growth charts over time to be sure that children have suitable body mass indices and that they do not exhibit changes in growth rate that are not acceptable for their ages. Nowhere do they state the conclusion that Ms.McIlwain seems to have reached.
There have been a number of empirical studies that have examined the connections (if any) between autism and gastrointestinal problems. One of these, a 2009 paper by S.H. Ibrahim et al. (Incidence of gastrointestinal symptoms in children with autism: A population-based study. Pediatrics, Vol. 124, pp. 680-686), reported differences between symptoms shown by persons diagnosed with autism and by non-autistic controls. These authors followed individuals until they were over 18 years of age, and examined the cumulative frequency of gastrointestinal problems-- in other words, they counted the number of people who had ever had a certain problem, not just the number who had it at a given time. They reported that their autistic group had a significantly higher cumulative frequency of constipation than the non-autistic group, with a frequency of 33.9% for the autistic group and of 17.6% for the comparison group. The autistic group also had a significantly higher frequency of food selectivity ("pickiness"), with this symptom occurring in 24.5% of the autistic group and 16.1% of the comparison group. There were no other significant differences in the overall count of gastrointestinal symptoms, nor in any other specific symptom. Of the symptoms that showed significant differences, neither was found exclusively in the autistic group, nor did even the majority of autistic children show the symptom in either case. As Ibrahim and his co-authors commented. "As constipation and... food selectivity often have a behavioral etiology, data suggest that a neurobehavioral rather than a primary organic gastrointestinal etiology may account for the higher incidence of these gastrointestinal symptoms in children with autism."
When it's impossible to do experimental work, as is the case for this topic, we fall back on "natural experiments" that compare two groups, or we examine statistical correlations between factors. These are valuable research methods, but they very often leave us in a quandary about which one of two factors caused the other. As Ibrahim and his co-authors suggested, when autism and GI symptoms go together, it may be the autistic characteristics that cause behaviors that lead to symptoms like constipation-- not the other way around.
In any case, looking at the original articles gives a very different view than what we'd get if we restricted our information to the report from the "Age of Autism" web site. As so often occurs, we find that simple answers to complicated questions are not the right answers.
