Autism Revisited

Is the autism epidemic due to more awareness and better diagnosis, or to diagnostic substitution? Is it due to something that has changed in our environment? Is there really an autism epidemic? Gil Eyal et al., the authors of the book The Autism Matrix: The Social Origins of the Autism Epidemic wonder if we have been asking the wrong questions. They think a more profitable question is “Why was autism rare in the past?” They think a historical and sociologic approach reveals the answer, and they think deinstitutionalization is a key factor.

They argue that “our practices for representing and intervening in childhood disorders are no longer constrained by the opposition between retardation and illness, but proceed as if they can ignore it.”  The boundary between mental retardation and mental illness has been blurred. Autism is defined as a mental disorder in the Diagnostic and Statistical Manual of Mental Disorders (DSM), but psychiatry and biomedicine have no effective medical treatments to offer; treatment has fallen into the hands of paraprofessionals, educators, and parents, and many of the treatments they use were originally developed to treat mental retardation or brain injuries. One whole institutional matrix of practices has given way to another.

The diagnostic criteria are a mess. Children in the autism spectrum demonstrate deficits in social interaction, verbal and nonverbal communication, and repetitive behaviors and interests. About a third of them are also diagnosed with mental retardation, but those on the Asperger’s end of the spectrum may have above-average IQs, and there are “idiot savants.” Many are concurrently diagnosed with epilepsy, hyperactivity, emotional disorders, and learning disability. The heterogeneity makes research difficult. In DSM-II autism was classified under childhood schizophrenia. In DSM-III (1980) it was classified under Pervasive Developmental Disorders. The diagnostic criteria were relaxed, and in DSM-IV (1994) head-banging and self-mutilation were no longer mentioned. In DSM-5 (2013) the diagnosis of “autism spectrum disorders” included autism, Asperger syndrome, pervasive developmental disorder not otherwise specified (PDD-NOS), and childhood disintegrative disorder. Rather than using science to direct diagnosis and clinical practice, DSM simply reflected current practice. The chief of child psychiatry at the National Institutes of Mental Health said, “I’ll call a kid a zebra if it will get him the educational services he needs.”

Autism has become protean. It can mean one thing and its opposite at the same time: profound mental retardation and near-genius abilities, hyper-sensitivity and hypo-sensitivity, aloofness and over-attachment, flat affect and combustible tantrums. Autism can be considered a developmental disorder positioned somewhere between mental illness and retardation. There is a “window” in which interventions might influence neural pathways; no such window exists in mental retardation. Autism provides flexibility for experts to postpone the medical diagnosis of social destiny (whether the child will function independently in society).

Some observers rejected the idea of autism as a unitary disease entity and saw it as “overlapping, concurrent, and deeply intertwined with all the other developmental disabilities – mental retardation, cerebral palsy, epilepsy, speech and language delays, learning disabilities.” Others thought autism was a symptom, not a syndrome.

The diagnosis of mental retardation had nothing to do with IQ. It was a legal and social concept that justified locking up those who were “socially incapable,” some of whom had higher IQs than some socially capable individuals. Parents were glad to have their children classified as autistic because there was less stigma attached to that diagnosis.

The book investigates the puzzle of variation in autism rates. The rate of designation of autism by school boards to provide services is 22 times higher in Maine than in Oklahoma. In Norway, the prevalence is 5.2 in 10,000; across the border in Sweden, it is 72.6 in 10,000. In Oregon, children are qualified as autistic for special education services if they simply demonstrate “impairments in social interaction.” Two questions must be answered: why is autism increasing but not learning disabilities or other disorders, and why now? After autism was named in 1943, it took 48 years for the numbers to start rising. As autism numbers rose, special education enrollments showed a corresponding decrease in mentally retarded or developmentally delayed; so there was clear evidence of diagnostic substitution.

For many years, doctors encouraged parents to commit very young children to institutions for the mentally retarded, out of concern for the well-being of the middle-class family. Eventually there was a push-back against the establishment doctors, as society recognized the value of retarded existence and came to believe that the middle-class family, in partnership with the state, was best for early detection and intervention. Parents became activists and children were de-institutionalized. The goal became “normalization” of behavior, with family, residential, educational, and work arrangements adapted to the needs and abilities of the individual child, whether mentally retarded or autistic or something else entirely. Parents became co-therapists, and a new network of (supposed) expertise emerged.

Early in the history of autism, parents were blamed as “refrigerator mothers.” Then the stress of raising an autistic child was recognized and parents were treated with psychotherapy. Then parents were considered experts who best understood their child and could translate for them; they assumed the role of therapists.

Autism therapies upset the boundaries between established and alternative medicine, between illness and retardation, and between expert and layperson. Defeat Autism Now! (DAN!) and similar organizations weaved in and out of science, went back and forth between medicine, alternative medicine, and the market. They blended therapy, parenting, and advocacy, and they glued new symptoms onto the autism spectrum.

Many parents are “therapy omnivores,” eclectically combining whatever they think works for their child. An Internet survey identified 111 different therapies. “Speech therapy was the most commonly reported intervention, followed by visual schedules, sensory integration, and applied behavior analysis. In addition, 52% of parents were currently using at least one medication to treat their child, 27% were implementing special diets, and 43% were using vitamin supplements.” There is little or no scientific evidence to support most of the treatments being used. Rather than controlled studies, they rely on collecting “n of 1” reports from parents. This potluck approach undermines science. Randomized, double-blind, placebo-controlled studies are the gold standard; among autism activists, scientific rigor is subordinated to the “silver standard” of clinical efficacy (= testimonials). That’s a dangerous approach, because development is delayed, not absent. These children tend to improve over time, and it’s easy to falsely attribute the improvement to whatever treatment they tried last. Even well-educated, highly intelligent people have succumbed to the post hoc ergo propter hoc fallacy. The account by James Laidler is well worth reading. He and his wife were both physicians, yet they managed to thoroughly fool themselves about the dietary and other treatments they used on their two autistic sons.

The book doesn’t deny the existence of autism, but it says, “the recent rise in autism diagnoses represents not an epidemic, but a change in the institutional conditions under which we perceive and treat childhood disorders…The very category of autism emerged in the interstice between mental retardation and mental illness.”

Conclusion: diagnosis and treatment of autism are deeply flawed

Whether or not you accept the premises of the book, the categorization of autism is blurry, as evidenced by the constant changes in each subsequent iteration of the DSM. Perhaps it would be better to avoid diagnostic labels entirely and to simply ask what each child’s handicaps and skills are and what we can do to help them. But unfortunately we rely on labels to authorize the provision of special educational and other services.

The book is full of facts and ideas that are well worth thinking about. But it is not an easy read. I found it dense, repetitive, slow going, and sometimes I struggled to understand. I could not recommend it for most readers, but it would be valuable for those with a special interest in autism.

This article was originally published in the Science-Based Medicine Blog.

Dr. Hall is a contributing editor to both Skeptic magazine and the Skeptical Inquirer. She is a weekly contributor to the Science-Based Medicine Blog and is one of its editors. She has also contributed to Quackwatch and to a number of other respected journals and publications. She is the author of Women Aren’t Supposed to Fly: The Memoirs of a Female Flight Surgeon and co-author of the textbook, Consumer Health: A Guide to Intelligent Decisions.