Autism: A History of Diagnostic Changes

Leo Kanner usually gets the credit for discovering autism in the 1940’s, but Russian psychiatrist Grunya Sukhareva described it 20 years before him

Autism has been misclassified, misunderstood, and redefined for nearly three centuries, and we still don’t have it right

STUDY: Pierucka M et al, History of Psychology 2026;29(1):20–30

STUDY TYPE: Review

FUNDING: Independent

Background

The paper traces the evolution of autism from:

• Before 1980: A rare childhood disorder with narrow criteria
• 1980-2013: Broader category with multiple subtypes, in children and adults
• 2013-present: A spectrum encompassing a wide range of conditions

Early Roots: Before DSM-III

1749

In one of the earliest descriptions, patient HB had impairments in language, eye contact, and social relationships. He was lacked tact and engaged in repetitive behaviors, always sitting in the same seat in church. His diagnosis with “silent madness” prompted legal troubles: his marriage was annulled, and his inheritance was lost to his brother.

1911

Eugen Bleuler coins the term “autism”, but as a feature of schizophrenia (withdrawal into inner life).

1920s

Grunya Sukhareva describes children with autistic traits using the term “schizoid psychopathy.”

1943

Leo Kanner publishes the first formal description of “infantile autism.” He distinguishes autism from schizophrenia by its early onset, “extreme aloneness,” and insistence on sameness. His descriptions resemble those of Dr. Sukhareva, but he does not reference her.

1944

Hans Asperger describes a milder form of autism, emphasizing social deficits with intact intelligence and restricted interests.

1960s–70s

Autism is seen as a reaction to cold, “refrigerator mothers,” and idea refuted by Rimland in his 1964 book Infantile Autism.

1970s

Research by Rutter, Kolvin, Chess establishes autism as separate from schizophrenia, with an earlier onset and unique neurological and comorbid features (eg, higher rates of epilepsy, something not seen in schizophrenia). Kolvin and colleagues looked at early-onset versus later-onset disorders and found that children with schizophrenia tended to have family histories of schizophrenia, while those with autism did not.

Official Recognition: DSM-III and Beyond

1980

DSM-III is first edition of DSM to recognize autism as a unique disorder, listed under “Pervasive Developmental Disorders.” They call it “infantile autism” with criteria emphasizing:

• • • Early onset (before 30 months)
    • Social, communication, and behavioral impairments
    • Explicit exclusion of schizophrenia features

The criteria were more narrow and rigid criteria than today’s, and focused on early childhood, with no adult description.

1987

DSM-III-R broadens the concept, removing “infantile” from the title and defining it with three domains:

• • • Social interaction
    • Communication
    • Restricted/repetitive behaviors

1994

DSM-IV expands the pervasive developmental category to include:

• • • Autistic disorder
    • Asperger’s disorder
    • Rett’s disorder
    • Childhood disintegrative disorder

2013

DSM-5 brings the pervasive developmental categories into a single diagnosis of autism spectrum disorder.

For the first time, this DSM allows autism and ADHD to be diagnosed together.

ICD-11 follows with a similar spectrum model, adding sensory processing issues, such as over- or under-responsiveness to sounds, sights, smells, tastes, or touch.

Looking Back

In 1988, Rutter and Schopler wrote that autism had the most validated diagnosis in child psychiatry. And yet, each subsequent revision made major changes that called that confidence into question. The history of autism is partly a history of psychiatry’s overconfidence — in our categories, our theories, and our ability to distinguish one presentation from another.

The prevalence has risen roughly twentyfold since the 1990’s, and we still don’t have a biomarker or criteria that everyone agrees on.

Practice Implications

Diagnostic criteria change, but one thing remains constant. Severity and functional impairment are what separates a disorder from a trait.