The Invisible Pandemic: How the World Slept Through the Largest Neurocrisis of the 21st Century

When the U.S. Centers for Disease Control and Prevention released its latest autism prevalence report in April 2025, the figure did more than update a spreadsheet. It demolished an illusion that bureaucrats, insurers, and politicians had nursed for decades: that autism is rare. One in 31 eight-year-olds. In California, one in 19. Among boys, one in 20. Health and Human Services Secretary Robert F. Kennedy Jr. called it "an epidemic raging out of control." The scientific community rejected the framing. It did not reject the number.

Chapter 1. The Arithmetic of an Epidemic

In 2002, the CDC's ADDM surveillance network recorded autism prevalence at 6.6 per thousand children — one in 150. By 2014, the figure had doubled to 1 in 59. By 2020, it reached 1 in 36. The latest report, drawing on 2022 data from 16 states, put the rate at 32.2 per thousand — 1 in 31. In two decades, the number had nearly quintupled.

The global picture confirms the trajectory. A Global Burden of Disease study published in The Lancet Psychiatry in February 2025 estimated 61.8 million people worldwide on the autism spectrum — 1 in every 127 human beings on the planet. Age-standardized prevalence stood at 788.3 per 100,000. Autism now ranks among the ten leading causes of non-fatal disease burden for people under 20.

Global averages, however, conceal dramatic disparities. In Japan, the GBD age-standardized prevalence reaches 1,587 per 100,000 — among the highest on record. South Korea follows at 1,279. In Israel, 28.3 per thousand children born in 2018 have received an autism diagnosis. In Russia, official statistics report 0.41 per thousand, but a population-based study published in 2025 found actual prevalence at 22.2 per thousand — an undercount of 54-fold.

Chapter 2. The Invisible Women

For decades, autism was considered a predominantly male condition. The 4:1 male-to-female ratio became textbook orthodoxy. A Swedish study published in the BMJ in February 2026 dismantled it. The authors tracked 2.7 million individuals born between 1985 and 2020 and found that by age 20, the diagnosed male-to-female ratio falls to 1.2:1. By adulthood, the gap is effectively indistinguishable.

The problem is not that fewer women are autistic. The problem is that no one sees them. Median age of diagnosis for boys in the United States is five. For girls, it is eight. One quarter of all female diagnoses in 2024 were made after the age of 19. Girls mask — they mimic social behavior, mirror the conversational patterns of their peers, and burn through their reserves until what emerges is depression, anxiety disorder, or in the worst cases, suicidal crisis.

The suicide data are unforgiving. Risk among autistic people runs two to nine times higher than the general population. 34 percent report suicidal ideation. In 2021, an estimated 13,400 excess suicide deaths were recorded among people on the spectrum globally. Women are disproportionately vulnerable — precisely because diagnosis and support arrive too late.

Chapter 3. A Lifetime in the Queue

In England, 254,108 people are on the waiting list for autism assessment. That is the figure as of December 2025. The National Institute for Health and Care Excellence recommends initial evaluation within 13 weeks. Fewer than 5 percent of referrals meet that standard. In Oxfordshire, the waiting list has reached 18 years; the service has stopped accepting new patients. In Scotland, 42,000 children are waiting for neurodevelopmental assessment. Some have been waiting up to ten years.

NHS England estimates that 750,000 people in the country live with undiagnosed autism. The total number of autistic individuals may exceed 1.2 million. The government's 2021–2026 autism strategy set six priority areas — from awareness to employment. Results at the time of publication remain modest.

Chapter 4. The Price of Silence

The lifetime cost of supporting one person with autism and co-occurring intellectual disability in the United States is estimated at $2.4 million. Without intellectual disability, $1.4 million. The aggregate annual cost of autism to the American economy in 2025 stands at $461 billion. Projections put the figure at $5.54 trillion by 2060.

Applied Behavior Analysis — ABA — is the standard intervention. It costs $40,000 to $60,000 per child per year. In North Carolina, state spending on ABA rose from $122 million in fiscal year 2022 to a projected $639 million in 2026 — an increase of 423 percent. In Nebraska, the jump was 1,700 percent. In Indiana, 2,800 percent. Bloomberg in February 2025 called ABA "one of the fastest-growing bills in American healthcare."

The paradox is this: ABA is also the most controversial method within the autistic community itself. A 2018 study found an 86 percent increase in PTSD symptoms among those who underwent intensive ABA. Only 5 percent of autistic adults endorse the therapy. Ireland in 2023 classified certain forms of ABA as equivalent to conversion therapy.

Chapter 5. Why the Numbers Rise

The scientific consensus: most of the growth reflects non-etiological factors — broadened diagnostic criteria, rising awareness, declining stigma. A 2015 Danish study showed that 60 percent of the increase among children born between 1980 and 1991 could be explained by changes in diagnostic practice alone. DSM-5, published in 2013, collapsed autism subtypes — Asperger's syndrome, PDD-NOS — into a single spectrum, widening the diagnostic net. ICD-11 in 2022 followed suit.

Genetics account for up to 80 percent of risk. Concordance among identical twins runs 60 to 90 percent. By 2024, more than 290 genes had been identified in association with autism. A Princeton/Simons Foundation study published in 2025 delineated four biologically distinct autism subtypes, each with a different genetic profile. Long-read genome sequencing reported in Cell Genomics in 2026 uncovered complex structural variants invisible to standard methods — potentially doubling the share of explainable heritability.

Environmental factors are not primary, but they are not zero. Fine particulate matter — PM2.5 — during the third trimester of pregnancy elevates risk. Parental age is a documented factor: fathers aged 45 to 46 carry 20 percent higher risk than those aged 23 to 24. Vaccines as a cause have been refuted repeatedly and conclusively. Studies involving millions of children found no association.

Chapter 6. The Political War

In April 2025, HHS Secretary Robert F. Kennedy Jr. announced a federal investigation into the causes of autism and promised answers by September. By March 2026, no answers have materialized. What materialized instead: $80 million in autism research cut through DOGE; NIH funding for autism down 26 percent between January and April 2025; a new Interagency Autism Coordinating Committee, appointed by Kennedy in January 2026, that includes anti-vaccine activists and excludes leading researchers.

Only 9 percent of the NIH autism budget goes to services research — the work that actually helps people live. 64 percent goes to biology and etiology. The disproportion, according to several experts, reflects an institutional priority: knowing "why" matters more than knowing "how to help."

Chapter 7. The Labor Market Is Closed

In the United Kingdom, the employment rate for autistic people is 22 percent — the lowest of any disability category. Globally, up to 85 percent of autistic adults are unemployed or underemployed. This is not a consequence of incapacity. It is a consequence of environments that were never built for neurodivergence. Sensory overload in open-plan offices, inflexible interview formats, absence of accommodations — these are systemic barriers, not individual limitations.

Chapter 8. The Science at the Frontier

SPARK, the Simons Foundation's research cohort, is the largest of its kind. It includes more than 100,000 autistic participants. By October 2024, 21,532 individuals had been sequenced; 8.6 percent received a genetic finding. Girls were more likely to carry a pathogenic variant — 12 percent versus 8 percent for boys.

Artificial intelligence is entering the diagnostic pipeline. Deep-learning video analysis can now detect behavioral markers of autism from recordings under one minute long, according to a study in Nature npj Digital Medicine published in October 2025. Multimodal systems that integrate audio interaction data with screening questionnaires achieve 79.6 percent accuracy in risk stratification.

The projections are sobering. By 2045, the global autistic population will reach 71.8 million. Among those over 70, the number will grow from 2.48 million in 2021 to 5.15 million by 2040. Disability-adjusted life years will rise 59 percent by 2030. No healthcare system on earth is built for that scale.

Chapter 9. The War of Narratives

In March 2026, Professor Uta Frith of University College London stated that the autism spectrum had become "so broad it is close to collapse" and had "lost meaning." She drew a line between "classic" autism — early diagnosis, intellectual disability — and a growing cohort diagnosed in adolescence and adulthood, often women. The autistic community and many researchers rejected the position.

CDC data show that roughly 13 percent of children diagnosed with autism subsequently lose the diagnosis. A Benford's Law analysis published in Frontiers in Psychiatry in 2025 found statistical anomalies in global prevalence data, particularly after the introduction of DSM-5. The counterargument from British researchers: a 747 percent rise in UK diagnoses between 1998 and 2018 reflects deliberate expansion of criteria — "there is no evidence that diagnoses are being made unnecessarily." With 750,000 undiagnosed people in England alone, underdiagnosis remains a far larger problem than overdiagnosis.

Chapter 10. What Comes Next

Autism is not an epidemic. It is not a trendy diagnosis. It is a neurobiological reality that the world has finally begun to see — and for which it is profoundly unprepared. Healthcare systems cannot absorb the queues. Schools are not adapted. The labor market is sealed shut. Research is underfunded and politicized. Families pay tens of thousands of dollars a year out of pocket.

No government has a strategy designed for a world in which 72 million people will require lifelong support by 2045. No insurance system is prepared for ABA therapy growing faster than any other line item. No education system knows how to include one student in 31 without lowering standards for the other thirty.

The invisible pandemic — not because it is quiet. Because the world prefers not to look.

ATLAS covers the stories that shape what you eat, earn, and breathe. If you want to know first — find us where the news moves fastest.