For many years, articles about autism cited prevalence rates of approximately 7 in 10,000.1 Over the past few years, however, there appears to have been an explosion in the rate at which autism is diagnosed: More recent estimates range from about 30 in 10,0002 to one in 68.3 References to an autism epidemic appear to have originated in a 2002 California legislative report suggesting a 273% increase in autism from 1987 to 1998.4
Concerns about rising rates of autism, however, are not new. In 1943, Leo Kanner, MD, a psychiatrist and pioneer in the study of autism, published a paper titled, “Autistic disturbances of affective contact.”5 The result? “Almost overnight, the country seemed to be populated by a multitude of autistic children,” he later observed.6
To what should we attribute the current rise in reported autism rates? Even a casual review of the autism literature suggests a number of potential causes that may account for at least a portion of the recent increase.
We know more about the disorder
In his paper, Kanner described the “peculiarities” of 11 children whom he had cared for.5 While several had been diagnosed with mental retardation, childhood schizophrenia, or both, what stood out to Kanner was an “autistic aloneness” evident from the beginning of life. This was in contrast to childhood schizophrenia, in which a child experienced a departure from previously normal interrelations. His insight contributed to our understanding of what is now recognized as autism spectrum disorder (ASD).
Also in 1943, Hans Asperger, MD, was studying families with children exhibiting behaviors similar to those described by Kanner. The following year, Asperger published an article (in German) describing these children. Unfortunately, this paper—titled “Autistic psychopathy in childhood”7—was not translated into English until the early 1990s.8
Since Kanner and Asperger first called attention to the disorder, there have been numerous changes in societal and medical understanding of autism. Bruno Bettelheim, PhD, an Austrian-born child psychologist with a particular interest in emotionally disturbed children, theorized that poor maternal bonding and lack of maternal affection were responsible for autistic characteristics.9 Bernard Rimland, PhD, a psychologist and the father of an autistic child, argued for a biological basis of the disorder.10 Rimland’s theory of autism as a neurodevelopmental disorder with an unidentified organic etiology is most consistent with current medical opinion.
Definitions and diagnostic criteria have evolved
The first edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM) was published in 1952.11 Although it was nearly a decade after Dr. Kanner clearly described autism as an entity separate from childhood schizophrenia, the word autism was used in this edition only once—to describe psychotic reactions associated with schizophrenia. Similarly, autism was referred to only in relation to childhood schizophrenia in the DSMII,12 published in 1968.
DSM-III adds diagnostic criteria. In DSM-III (1980),13 specific diagnostic criteria for infantile autism and pervasive developmental disorder (PDD) appeared for the first time. Both diagnoses were clearly contrasted with a diagnosis of schizophrenia, and were used to identify children who exhibited a pervasive lack of responsiveness to others.
The DSM-III-Revised (R) (1987)14 added a classification scheme more consistent with the current standard. It included: 1) qualitative impairment in reciprocal social interaction; 2) qualitative impairment in communication and imaginative activity; and 3) restricted activity/interests.
PDD Not Otherwise Specified was also included, and served to identify those who had qualitative impairments in social interaction and communication skills but did not meet the full criteria for autism disorder or PDD.
The publication of DSM-IV (1994)15 brought another change: Specific criteria were outlined for the diagnoses of Asperger syndrome, Rett syndrome, and childhood disintegrative disorders; in the DSM-IV-Text Revision (TR) (2000),16 this structure remained relatively stable.
DSM-5 (2013) took another step, consolidating these various disorders into a single, unifying diagnosis of ASD.17 Significant controversy surrounded this change, with some viewing it as an oversimplification that does not accurately reflect important distinctions among divergent disorders18 and others arguing that it will result in unrecognized cases and exclusion of affected individuals.19 Notably, a recent study using concurrent DSM-IV and DSM-5 criteria to diagnose autism and PDD in more than 4000 children documented a high level of agreement between them.20
Diagnostic tools have improved
As the number of children diagnosed with autism has increased, so have efforts to more accurately diagnose autism as a distinct disorder.
In the 1960s, guidelines for diagnosing autism focused primarily on Kanner’s original descriptive criteria. Even in the 1980s, after DSM-III criteria identified autism as a distinct disorder, children were being evaluated with generalized developmental screening tools focused on behaviors characteristic of a severe mental handicap, without differentiating between autistic and nonautistic children.