How Many People Are Autistic? Why Estimates Range from Under 1% to Over 3%

Around the world, an estimated 1 in 127 people are autistic. Ask country by country, though, and published figures run from under 1% to over 3% of children — and the research suggests that most of that spread reflects how hard each place looks for autistic people, not how many autistic people live there.

A prevalence figure sounds like a simple fact about a population. In practice it is also a record of who has been noticed. A low number rarely means few autistic people; more often, the studies below suggest, it marks a shortage of assessment services, of clinicians who recognise autism in girls and adults, of school systems that record support needs, or of anyone asking at all. The child in a mainstream classroom who has never been assessed, the older adult who grew up before autism was widely understood, the family whose nearest diagnostic service is out of reach: all of them belong in the true figure, and many are missing from the official one. Services are planned from the numbers, and whoever the count leaves out, the budget usually leaves out too.

Roughly 1 in 100 worldwide — and even that depends on method

The GBD 2021 study, which models every country, put global age-standardised prevalence at 788.3 per 100,000 people in 2021 — just under 0.8% — with 1,064.7 per 100,000 among males and 508.1 per 100,000 among females^[1]. The World Health Organization uses the same headline figure and adds two caveats in the same breath: the estimate “represents an average figure, and reported prevalence varies substantially across studies”, and “some well-controlled studies have, however, reported figures that are substantially higher”^[2].

Even the global number moves with method. The previous GBD round put autism at 1 in 271 people. The authors attribute the jump to 1 in 127 mainly to a change in their own approach: they dropped studies relying on “passive case finding (eg, registry or administrative prevalence estimates) that probably underestimated the prevalence of the autism spectrum”^[1]. The estimate more than doubled between editions largely because the counting changed, not the world.

Reviews of the underlying studies agree. A 2022 systematic review of 99 estimates from 34 countries found a median of 100 per 10,000 (1%), but a range from 1.09 to 436 per 10,000 — roughly 0.01% to 4.4%^[3]. A 2023 meta-analysis of studies from 29 countries, covering 1994 to 2019, pooled 0.72%^[4]. Both are averages across very different studies.

Country by country, estimates run from under 0.5% to over 4%

The chart below gathers published estimates from twelve countries alongside the global model, and the table explains how each was produced. The highest figures come from whole-population screening, thorough records review or household surveys; the lowest mostly from diagnosis registers, plus an eight-city screening study in China. None of these numbers means much without its age group, year and method. They are not directly comparable with one another, and a taller bar does not mean a country has more autistic people.

Within single countries the spread is just as wide. Across the 16 US sites, prevalence ran from 9.7 per 1,000 in Laredo, Texas, to 53.1 in California — more than fivefold^[7]. Across the Canadian provinces and territories covered, it ran from 1.33% in Saskatchewan to 2.97% in Prince Edward Island^[13]. Iceland’s figure was about four times Finland’s, for children of the same age in the same year, in two Nordic countries with universal health care. The researchers concluded that the differences “likely reflect variation in detection, referral and diagnosis practices and autism awareness across these areas”^[8]. Even the GBD model gives national figures from 588.2 per 100,000 in Bangladesh to 1,586.9 in Japan^[1].

How autistic people are counted shapes the answer

The 2023 meta-analysis compared study designs directly. Studies in which trained reviewers searched health and education records found a pooled prevalence of 1.22%; studies built on health-insurance data found 0.35%, with registers, administrative databases and direct surveillance in between^[4].

Each method has its own blind spot. Canada’s public health agency notes that its health-system surveillance can miss children diagnosed “through private healthcare (e.g., private psychology clinics), the education system, or research institutions”^[13]. In the same country, a parent survey put diagnosed autism among 1–17-year-olds at 3.0% in 2023, against 2.25% in health records for 1–19-year-olds in 2023–24^[9],[13].

The US monitoring network is explicit: “Research has not demonstrated that living in certain communities puts children at greater risk for developing ASD.” Differences between sites “might be due to differences in availability of services for early detection and evaluation and diagnostic practices” — California, the highest site, has trained hundreds of local paediatricians to screen and refer early^[7].

Active screening, meanwhile, keeps finding people no service had seen. In the South Korean study, “two-thirds of ASD cases in the overall sample were in the mainstream school population, undiagnosed and untreated”^[11]. In China’s eight-city study, 43.3% of the autistic children identified were newly diagnosed, and 90.4% of those were attending regular schools^[14]. Even some wealthy countries have no population count. The OECD reports that “Germany has no available data on the population prevalence of autism or rates of diagnosis”; the national figure it could compare, based on hospital inpatient diagnoses, was 5.5 per 100,000 people in 2023^[15].

Poorer countries record less autism — and many have no count at all

Measured prevalence also tracks national income. In the 2023 meta-analysis, pooled prevalence was 0.79% in high-income countries and 0.32% in lower-middle-income countries; low-income countries were excluded from that comparison^[4]. Of the 317 estimates in the review, just one came from Africa and four from South America^[4]. The authors suggest the income difference “could be attributed to the detection gap, associated with lower access to appropriate diagnostics, lower awareness among parental and professional communities about autism, and specific cultural attitudes” that shape whether families seek care^[4].

The WHO is blunt: “The prevalence of autism in many low- and middle-income countries is unknown”^[2]. Where careful studies have been done, autism does not look rarer. A 2026 systematic review found only seven studies from four African countries — Egypt, Kenya, Uganda and Nigeria — with 71,341 participants between them. Pooled, clinically confirmed prevalence was 1%, with very wide variation between studies, and the authors concluded that African estimates “appear comparable to those reported in high-income countries”^[16]. A 2011–12 study of 3,964 children aged 2–9 in five Indian districts likewise found autism prevalence “comparable to the published global literature”^[17]. The limited evidence suggests the income gap is largely a gap in finding people — see the global detection gap.

Counts have climbed steeply — mostly, the evidence suggests, because recognition has

Almost everywhere with a time series, recorded autism has risen. In the US monitoring network, the figure for 8-year-olds went from about 1 in 150 in 2000 to 1 in 31 in 2022^[18]. Canada’s health-system surveillance went from 1 in 714 children and youth in 2000–2001 to 1 in 44 in 2023–2024^[13]. In UK primary care, recorded new diagnoses rose 787% between 1998 and 2018, with the greatest rises among adults and a larger increase for females than males^[19]. The OECD found most countries it reviewed had seen average annual increases of 6–10% in autism detection among children over the past decade^[15]. Pooled across studies, prevalence was 0.25% for 1994–1999 and 0.99% for 2015–2019 — though these are different studies, not repeat counts^[4].

Why? The prevailing explanation is changing recognition rather than a change in how many people are autistic — an explanation with good evidence behind it, but not a settled fact. In Denmark, a change in diagnostic criteria in 1994 and the inclusion of outpatient diagnoses in 1995 together accounted for about 60% of the rise in recorded autism among children born 1980–1991^[20]. In Sweden, among children born 1993–2002, the prevalence of autism traits measured by parent interview in a large twin study “was stable”, while registered diagnoses rose substantially^[21]. The UK researchers judged that increases “could be due to growth in prevalence or, more likely, increased reporting and application of diagnosis”^[19], and the 2023 meta-analysis concluded that rising estimates “probably are not caused by an increase in true prevalence”^[4]. The OECD is more cautious still: “Currently, it is unclear” whether rising diagnoses reflect a real increase or more people being diagnosed^[15]. Either way, each new count reveals people who need services sized for a higher number.

The people most likely to be missing are adults

Almost every national count is a count of children. In England’s GP records in 2018, 2.94% of 10–14-year-olds had an autism diagnosis (1 in 34), against 0.02% of people aged 70 and over (1 in 6,000)^[10]. As the authors put it, many autistic adults “missed out on a diagnosis as children when autism was little known”. The same study’s exploratory projections suggest that 463,500 people in England had been diagnosed by 2018, and that between 435,700 and 1,197,300 more may be autistic and undiagnosed^[10].

Direct adult measures are scarce. A study combining clinical assessments from England’s Adult Psychiatric Morbidity Survey with a survey of adults with intellectual disability estimated that 1.1% of adults were autistic (95% confidence interval 0.3% to 1.9%)^[22]. The survey’s 2023/24 autism findings were withdrawn on 18 May 2026 after a review found a data error; further investigation then identified “limitations in the validity of the autism assessments”, and NHS England says the values are “not sufficiently robust to support a reliable estimate”. A revised chapter is expected in autumn 2026^[23]. In the US, where no surveillance system had been funded to count adults, researchers modelled 2.21% of adults — about 5.4 million people — as autistic in 2017, diagnosed or not^[24]. Canada’s surveillance system states that “no case definition could reliably identify Autistic adults in health administrative data”^[13]. More in diagnosed in adulthood.

Counts also miss people unevenly. In the US sites, autism was identified 3.4 times as often in boys as in girls^[7] (see autistic women and girls). In 2022, prevalence was higher among Asian or Pacific Islander, Black and Hispanic children than White children^[7], although until 2016 White children had the highest estimated prevalence^[15] (see who gets identified).

What the evidence asks of us

If a low number usually means autistic people have not been found, counting is the first step in deciding who gets support. The evidence points to five things governments, services and funders can do.

1. Count properly, and keep counting. The WHO–UNICEF global report on developmental disabilities names as one of its action areas the need to “strengthen health information systems, monitoring of programmes and services, and research for data-driven decisions and accountability”^[25].
2. Count adults. England’s GP-record study found “an urgent need to improve access to adult autism diagnostic services”^[10], and a robust, reissued adult survey estimate would help plan them^[23]. US researchers built their adult estimate so that states could “estimate the need for diagnosing and providing services to those unidentified”^[24].
3. Invest where the data are thinnest. With prevalence unknown in many low- and middle-income countries^[2], the African review calls for urgent action to “expand diagnostic confirmation services and strengthen training for primary healthcare professionals”^[16].
4. Plan for recognition to keep rising. The CDC says increased identification, “particularly among very young children and previously underidentified groups, underscores the increased demand and ongoing need for enhanced planning to provide equitable diagnostic, treatment, and support services”^[7]. The OECD found that “capacity in public diagnostic systems has not always kept pace with rising demand”^[15].
5. Publish every number with its method. A figure without its age group, year and method invites misleading league tables. Where the evidence points to better recognition, a rising count is a sign of people finally being seen, and a reason to fund support rather than a cause for alarm.

This article is part of our Autism & Health research library.

Language: this piece uses identity-first language (“autistic people”), the preference of most autistic-led organisations, and uses clinical terms such as “autism spectrum disorder” (ASD) only when naming a study, dataset or diagnostic category.

Why we're publishing this

Health Insurance UK is a commercial health-insurance resource, not a charity and not a campaign. We put this page together because “how many autistic people are there?” is one of the first questions any autism organisation is asked — and the honest answer is more useful than a single number. The case these figures make belongs to autistic people and the organisations working alongside them. Our part is to set the numbers out with their methods attached and say plainly what they show: where autistic people are looked for carefully, they are found, and the places with the lowest counts are usually the places with the least support.

How to read this data

Prevalence here means the share of a population who are autistic at a point in time. The sources are not interchangeable. Screening studies (Japan, South Korea, China) screen a whole population and assess those flagged, so they find undiagnosed people, but each covers a defined area (one community in Japan and in South Korea, eight cities in China) and may adjust for non-response. Records-based surveillance (the US ADDM Network) finds only children who have reached a service. Registers and administrative data (the Nordic countries, Canada’s health-system surveillance, England’s GP records) count recorded diagnoses. Surveys (Australia, Canada’s child health survey) rely on what people or parents report. The GBD figure is a statistical model that estimates every country, drawing on data from the 33 that have it. Most estimates are for children. Figures from different studies are not a trend, and the US series changed sites over time. The “more than fivefold” and “about four times” comparisons are our calculations from the cited figures. This is a summary of published data, not new research.

Use this data

Free to cite with attribution to Health Insurance UK. Charities, campaigners and journalists are welcome to reuse these figures and charts — please keep each figure with its age group, year and method. For the underlying figures as a spreadsheet, get in touch.

Important: This article summarises published research for general information. It is not medical advice and cannot be used to diagnose, treat or predict anything about an individual. If you have concerns about your own or someone else’s health, please speak to a qualified health professional.

Sources

1. GBD 2021 Autism Spectrum Collaborators (Santomauro DF, Erskine HE, Mantilla Herrera AM, et al.). The global epidemiology and health burden of the autism spectrum: findings from the Global Burden of Disease Study 2021. The Lancet Psychiatry 12(2):111–121, 2025. europepmc.org/article/MED/39709974
2. World Health Organization. Autism (fact sheet), updated 17 September 2025. WHO, 2025. who.int/news-room/fact-sheets/detail/autism-spectrum-disorders
3. Zeidan J, et al. Global prevalence of autism: a systematic review update. Autism Research, 2022. europepmc.org/article/PMC/PMC9310578
4. Talantseva OI, et al. The global prevalence of autism spectrum disorder: a three-level meta-analysis. Frontiers in Psychiatry, 2023. frontiersin.org/journals/psychiatry/articles/10.3389/fpsyt.2023.107…
5. Australian Bureau of Statistics. Autism in Australia, 2022 (Survey of Disability, Ageing and Carers). ABS, released 11 October 2024. abs.gov.au/articles/autism-australia-2022
6. Saito M, Hirota T, Sakamoto Y, et al. Prevalence and cumulative incidence of autism spectrum disorders and the patterns of co-occurring neurodevelopmental disorders in a total population sample of 5-year-old children. Molecular Autism 11:35, 2020. europepmc.org/article/MED/32410700
7. Shaw KA, Williams S, Patrick ME, et al. Prevalence and early identification of autism spectrum disorder among children aged 4 and 8 years — Autism and Developmental Disabilities Monitoring Network, 16 sites, United States, 2022. MMWR Surveillance Summaries 74(SS-2):1–22, Centers for Disease Control and Prevention, 2025. cdc.gov/mmwr/volumes/74/ss/ss7402a1.htm
8. Delobel-Ayoub M, Saemundsen E, Gissler M, et al. Prevalence of autism spectrum disorder in 7–9-year-old children in Denmark, Finland, France and Iceland: a population-based registries approach within the ASDEU project. Journal of Autism and Developmental Disorders 50:949–959, 2020. europepmc.org/article/MED/31813107
9. Public Health Agency of Canada. Prevalence: chronic conditions in childhood (2019 and 2023 Canadian Health Survey on Children and Youth). Health Infobase, Government of Canada. health-infobase.canada.ca/chronic-conditions/childhood/prevalence.html
10. O'Nions E, Petersen I, Buckman JEJ, et al. Autism in England: assessing underdiagnosis in a population-based cohort study of prospectively collected primary care data. The Lancet Regional Health – Europe 29:100626, 2023. europepmc.org/article/PMC/PMC10114511
11. Kim YS, Leventhal BL, Koh YJ, et al. Prevalence of autism spectrum disorders in a total population sample. American Journal of Psychiatry 168:904–912, 2011. europepmc.org/article/MED/21558103
12. Maciver D, et al. Prevalence of neurodevelopmental differences and autism in Scottish primary schools 2018–2022. Autism Research 16:2403–2414, 2023. europepmc.org/article/MED/38009513
13. Public Health Agency of Canada. Understanding autism data in the Canadian Chronic Disease Surveillance System. Health Infobase, Government of Canada, 23 December 2025. health-infobase.canada.ca/autism
14. Zhou H, Xu X, Yan W, et al. Prevalence of autism spectrum disorder in China: a nationwide multi-center population-based study among children aged 6 to 12 years. Neuroscience Bulletin 36:961–971, 2020. europepmc.org/article/MED/32607739
15. OECD. Policy Responses to Rising Autism Diagnoses in Childhood: Across the Spectrum. OECD Publishing, Paris, 2026. oecd.org/content/dam/oecd/en/publications/reports/2026/03/policy-re…
16. Ferreira IL, et al. Prevalence of autism in African countries: a systematic review and meta-analysis. Trends in Psychiatry and Psychotherapy, 2026. europepmc.org/article/MED/41637193
17. Arora NK, et al. Neurodevelopmental disorders in children aged 2–9 years: population-based burden estimates across five regions in India. PLOS Medicine 15(7):e1002615, 2018. journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1002615
18. Centers for Disease Control and Prevention. Data and statistics on autism spectrum disorder (ADDM Network prevalence by surveillance year, 2000–2022). CDC. cdc.gov/autism/data-research/index.html
19. Russell G, Stapley S, Newlove-Delgado T, et al. Time trends in autism diagnosis over 20 years: a UK population-based cohort study. Journal of Child Psychology and Psychiatry 63:674–682, 2022. europepmc.org/article/MED/34414570
20. Hansen SN, Schendel DE, Parner ET. Explaining the increase in the prevalence of autism spectrum disorders: the proportion attributable to changes in reporting practices. JAMA Pediatrics 169:56–62, 2015. europepmc.org/article/MED/25365033
21. Lundström S, Reichenberg A, Anckarsäter H, Lichtenstein P, Gillberg C. Autism phenotype versus registered diagnosis in Swedish children: prevalence trends over 10 years in general population samples. BMJ 350:h1961, 2015. europepmc.org/article/MED/25922345
22. Brugha TS, Spiers N, Bankart J, et al. Epidemiology of autism in adults across age groups and ability levels. British Journal of Psychiatry 209:498–503, 2016. europepmc.org/article/MED/27388569
23. NHS England. Adult Psychiatric Morbidity Survey: Survey of Mental Health and Wellbeing, England, 2023/4 (autism chapter notice). NHS England Digital, published 27 November 2025; notice dated 25 August 2026. digital.nhs.uk/data-and-information/publications/statistical/adult-…
24. Dietz PM, et al. National and state estimates of adults with autism spectrum disorder. Journal of Autism and Developmental Disorders 50:4258–4266, 2020. europepmc.org/article/MED/32390121
25. World Health Organization and UNICEF. Global report on children with developmental disabilities: from the margins to the mainstream. Executive summary. WHO and UNICEF, 2023. unicef.org/media/145021/file/Executive-Summary-Global-report-on-chi…