Autism spectrum disorder is not as certain as implied

Editor—Szatmari's editorial implies more certainty about the existence of an autism spectrum and causes of autism than is warranted.¹ Debate remains about the validity and usefulness of a broad definition of autism.²

Autism and Asperger's syndrome are distinct, and although they share common difficulties in social relatedness and obsessiveness, they can be distinguished in these.

In autism children are withdrawn, whereas in Asperger's syndrome they desire social contact but cannot negotiate rules.

In autism, which, unlike Asperger's syndrome, is associated with mental retardation, obsessiveness commonly entails routine behaviours and physical objects, whereas in Asperger's syndrome it entails idiosyncratic and often highly intellectual interests.

The distinction is comparable to that between schizophrenia and schizoid personality disorder. Like autism and Asperger's syndrome they share common features and genetic loading, but clinically their distinction is crucial. A broad definition risks confusion and overdiagnosis, an increasing problem when worried parents demand diagnostic certainty.

Szatmari's statement that autism is a neuropsychiatric disorder plays into this. Neurological problems have a bearing on autism, but the relation remains obscure and the implied claim that autism results from a primary neurological disorder is based on slim evidence. A genetic link does not necessarily imply neurological damage.

Szatmari acknowledges environmental factors but omits social and emotional factors, although the importance of these for psychological and brain development is well established. Evidence indicating their role in autism comes from studies of Romanian adoptees.³

Szatmari believes that research that supports organic causation helps temper parental guilt. In my experience, helping parents understand their unwarranted feelings of blame about their children is more productive than allaying anxiety with premature certainty about causation.