TL;DR
An analysis of CDC ADDM data from 2000–2016 found the increase in autism diagnoses was concentrated among children with milder adaptive challenges, while prevalence of moderate-to-profound cases declined slightly. Authors and commentators argue this pattern reflects broader identification and changing diagnostic thresholds rather than a population-level rise in severe autism.
What happened
Researchers analyzing Autism and Developmental Disabilities Monitoring (ADDM) Network records for 8-year-old children from 2000 through 2016 used adaptive behavior and cognitive scores abstracted from health and education files to track autism prevalence by adaptive-functioning level and co-occurring intellectual disability (ID). The study found that prevalence of autism with mild, borderline, or no significant adaptive challenges increased substantially, whereas prevalence of autism accompanied by moderate to profound adaptive challenges fell modestly. The growth in diagnosed cases was larger among children without co-occurring ID than among those with ID. Investigators and commentators interpret these results as showing that the apparent rise in autism diagnoses is confined to milder phenotypes—consistent with expanded identification or shifts in diagnostic standards—and note that improved therapies that raise adaptive or cognitive functioning could also have played a role.
Why it matters
- Public-health surveillance that does not separate severity levels can mask shifts in who is being diagnosed, affecting resource planning for services.
- If increases are driven by milder cases and diagnostic expansion, arguments that external factors (for example, vaccines or single exposures) caused a broad rise in autism are weakened.
- Education and clinical systems may need adjusted approaches to account for a broader, more heterogeneous autistic population with varied support needs.
- Understanding whether improvements in adaptive functioning are due to therapy versus diagnostic change affects policy on early intervention and funding priorities.
Key facts
- Source data come from the US Centers for Disease Control and Prevention’s Autism and Developmental Disabilities Monitoring (ADDM) Network for years 2000–2016.
- Prevalence of autism with mild, borderline, or no significant adaptive challenges rose from 5.1 per 1,000 (95% CI: 4.6–5.5) in 2000 to 17.6 per 1,000 (95% CI: 17.1–18.1) in 2016.
- Prevalence of autism with moderate to profound adaptive challenges decreased slightly from 1.5 per 1,000 (95% CI: 1.2–1.7) to 1.2 per 1,000 (95% CI: 1.1–1.4) over the same period.
- The increase in overall diagnosed autism was larger for cases without co-occurring intellectual disability than for cases with intellectual disability.
- The study used adaptive behavior and cognitive scores taken from existing health and educational records to categorize cases by adaptive level and ID status.
- Commentators cited in the piece argue the pattern indicates improved identification of milder forms or relaxed diagnostic standards; the authors also note the possibility that therapies improving functioning contributed to trends.
- The blog post references earlier work drawing similar conclusions from other datasets, including prior analyses based on Sweden and earlier CDC data.
What to watch next
- Whether diagnostic thresholds or official diagnostic criteria changed in ways that would explain the shift in case mix: not confirmed in the source
- Trends in autism prevalence and severity after 2016 to see if the pattern of growth among milder cases continued: not confirmed in the source
- Evidence quantifying how much improved therapies or interventions have altered adaptive and cognitive scores over time: not confirmed in the source
Quick glossary
- Autism: A neurodevelopmental condition characterized by differences in social communication and restricted or repetitive behaviors; presentations and support needs vary widely.
- Adaptive functioning: A person’s ability to manage daily life skills and independence, often assessed in educational and clinical settings to determine support needs.
- Intellectual disability (ID): A condition marked by limitations in intellectual functioning and adaptive behavior that appear during the developmental period.
- Prevalence: The number of existing cases of a condition in a specified population at a given time, often expressed per 1,000 or per 10,000 people.
- CDC ADDM Network: A US Centers for Disease Control and Prevention surveillance program that monitors autism and developmental disabilities in defined sites using medical and educational record review.
Reader FAQ
Did the study find autism has become more common overall?
The analysis found more diagnosed cases over time, but the increase was confined mainly to milder presentations; prevalence of moderate-to-profound cases did not rise and slightly declined.
Does this rule out vaccines or Tylenol as causes?
The source reports commentators concluding that because severe autism did not increase, claims that vaccines or Tylenol explain a broad rise in autism are unsupported; the study itself documents shifts in case severity rather than establishing causes.
What data did the researchers use?
They used adaptive behavior and cognitive scores abstracted from health and education records within the CDC’s ADDM surveillance network for children born in cohorts tracked between 2000 and 2016.
Are improvements from therapy responsible for the trends?
The authors mention that increased access to therapies that boost intellectual or adaptive functioning could have contributed, but the extent of such an effect is not established in the source.
Autism Hasn’t Increased by Alex Tabarrok January 1, 2026 at 7:18 am in Medicine Autism diagnoses have increased but only because of progressively weaker standards for what counts as autism. The…
Sources
- Autism hasn't increased
- Autism diagnoses are on the rise – but autism itself may not …
- Is There an Autism Epidemic? | Johns Hopkins
- Autism diagnosis on the rise, according to trends study
Related posts
- How Cameras and Lenses Capture Light: From Sensors to Pinhole
- Python Performance and Memory: Numbers Every Programmer Should Know
- Public domain 2026: Betty Boop, Pluto, and Nancy Drew set free