Introduction
In recent decades, the prevalence of Autism Spectrum Disorder (ASD) has been increasing exponentially in several regions of the world. Data from the CDC (USA) indicate that the rate rose from approximately 1:150 in 2000 to 1:36 in 2020, with recent estimates approaching 1:31 in some US states. This phenomenon has been the subject of intense debate. The most widely held interpretation suggests that prevalence increased due to increased detection and improved screening methods. However, a critical analysis reveals that the observed increase results from a multifactorial set: part real (biological), part administrative (changing criteria), and part artificial (diagnostic confusion).
1. Real Biological Contributions
1.1 Advanced Parental Age
There is consensus that advanced paternal and maternal age are associated with a higher risk of ASD and intellectual disability. This risk is mediated by the increase in de novo mutations, epigenetic alterations, and obstetric complications. As couples tend to postpone parenthood in contemporary societies, this demographic shift contributes to a real portion of the increase in prevalence.
1.2 ASD–ID Overlap
Another factor is the inclusion, in the current criteria, of children with intellectual disabilities (ID) and social/communication difficulties under the ASD umbrella. This overlap expands the population eligible for diagnosis, inflating the numbers without necessarily representing an increase in the “pure” incidence of ASD.
2. Changes in Diagnostic Criteria and Expansion of the Spectrum
2.1 From Categorization to Spectrum
The DSM-5 (2013) unified previously distinct categories—Asperger syndrome, PDD-NOS, and classic infantile autism—into a single spectrum. This increased diagnostic sensitivity, particularly in mild cases, and substantially increased administrative prevalence.
2.2 Educational Incentives and Convenience Diagnoses
In the North American educational system, diagnostic substitution has been observed: categories such as “speech/language impairment” have declined while “autism” has increased. This is explained by the fact that an ASD diagnosis guarantees greater access to specialized services than labels that are less administratively valued.
3. Diagnostic Confusion in Children
3.1 Isolated Speech Delay
Children with isolated speech delay, especially boys, are frequently misclassified as having ASD. Population screening tools, such as the M-CHAT, have high sensitivity but low specificity—resulting in many false positives, especially in early screening settings.
3.2 SES, Genetic Predisposition (EA4), and Screening
Children with a low polygenic score for educational achievement (EA4) and living in low socioeconomic status (SES) environments are more likely to have language delays. These contexts are associated with less verbal stimulation, greater socioeconomic pressure on parents, and the use of individual screens as a substitute for interaction. The resulting phenotype—speech delay without core social impairments—can be confused with ASD, artificially inflating the prevalence.
3.3 Impact on therapies and efficacy statistics
Many of these children, once diagnosed with ASD, are referred for early ABA therapy. Because many “late talkers” spontaneously progress to normal language development, their subsequent improvement can be mistakenly attributed to treatment. This mechanism generates both an inflated prevalence of ASD and an overestimation of the effectiveness of certain interventions in observational studies.
4. Diagnostic confusion in adults
The increase in ASD diagnoses in adults represents a distinct phenomenon. Many of these cases involve confusion between core ASD symptoms (present since childhood, with qualitative alterations in social reciprocity and restricted interests) and internalizing symptoms such as social anxiety, depression, or schizotypal traits.
Many adults seek an ASD diagnosis as an identity-based explanation for interpersonal difficulties, but this often involves not genuine autism but rather isolated or overlapping internalizing conditions. This contributes to inflated numbers in adult populations.
5. Integrated Model
: The increase in ASD rates can be understood as resulting from four main vectors:
1. Actual biological: advanced parental age, de novo mutations, and risk associated with ID.
2. Diagnostic expansion: DSM-5 changes, category unification, and diagnostic substitution.
3. Confusion in children: isolated speech delay in low-SES/AS4 contexts, often with the use of individual screens replacing parental interaction.
4. Confusion in adults: internalizing symptoms confused with core ASD.
Conclusion:
The so-called “autism boom” should not be interpreted as a homogeneous increase in incidence, nor as a mere statistical illusion. It is a multifactorial phenomenon: part genuine, linked to demographic changes and biological risks; part administrative, resulting from changes in diagnostic criteria; and part artificial, resulting from clinical confusion in both children and adults.
Recognizing these layers is essential to avoid overdiagnosis of non-ASD conditions, ensure appropriate interventions for different profiles, and improve epidemiological accuracy.
