Noticed at Two, Diagnosed at Five: Age at Autism Diagnosis Around the World

Around the world, parents usually first notice that their child is developing differently at about the age of two. The diagnosis tends to come two to three years later, and for many children much later than that. Across 30 studies from high-, middle- and low-income countries, parents first raised concerns at an average of 23.6 months, but diagnosis did not come until 56 months^[1].

The averages hide a familiar story. A parent raises a worry at a routine check-up and is told to wait and see. They raise it again, are referred, and wait, while their child starts school without anyone yet understanding how they experience the world. As the World Health Organization puts it, “Characteristics of autism may be detected in early childhood, but autism is often not diagnosed until much later”^[2].

Parents notice at around two, but diagnosis comes at around five

The broadest global estimate comes from a 2021 meta-analysis of studies published between 2012 and 2019. It drew on 56 studies that reported age at diagnosis for 40 countries and more than 120,000 autistic people. The pooled mean age at diagnosis was 60.48 months, just over five years. In studies of children aged 10 and under only, it was 43.18 months, about 3.6 years^[3]. An earlier review of studies from 1990 to 2012 found mean ages of 38 to 120 months, falling over time^[5].

The gap behind those averages matters more. A 2022 review brought together 30 studies, from the US, UK, Germany, Spain, Australia, Brazil, Colombia, India, Nepal, Malaysia and elsewhere, that recorded both when parents first became concerned and when diagnosis came. On average, first concerns came at 23.64 months and diagnosis at 55.97 months, a delay of 32.33 months^[1].

The review also split the studies by national income. In high-income countries, children were diagnosed at 52.60 months on average. In low- and middle-income countries, the average was 60.38 months. The delays were 28.67 and 37.12 months respectively. None of these differences was statistically significant^[1]. That does not prove every country does equally well, but no income group had closed the gap. Parents in richer and poorer countries alike began to worry at almost the same age (23.93 and 23.27 months), and then waited two to three years for an answer.

From first worry to diagnosis takes years

What families report shows the same pattern. In a survey of 2,520 caregivers in six Latin American and Caribbean countries, concerns began by 22 months, but diagnosis came at 46 months. The authors note that “children were diagnosed around the time they entered formal schooling, delaying the access to early intervention programs”^[6]. In a Nigerian clinic sample of 60 children, parents noticed the first signs at a mean age of 17.0 months. Diagnosis came at a mean of 9.00 years, 91 months after parents' first concern^[7]. That is one small clinic, not a national figure, but it shows how long the wait can be.

In a survey of 663 carers in Italy, Spain and the UK, concerns typically began when the child was 12 to 18 months old. In the UK, 49% waited more than a year just for a screening appointment, compared with 22% in Spain and 15% in Italy. Then 68% of UK carers waited more than another year from screening to a confirmed diagnosis, against 42% in Spain and 24% in Italy^[8]. An earlier UK survey of 1,047 parents found that diagnosis took around 3.5 years on average from when parents first went to a professional^[9]. In the US, parents of 584 Black (African American) autistic children reported an average age at diagnosis of 64.9 months, 42.3 months after their first concerns^[10]. Formal assessment queues are one part of this; see autism assessment waiting times around the world.

Where a child grows up changes when they are diagnosed

Countries measure age at diagnosis in different ways (medians or means, records or parents' reports, children only or all ages), so the figures below are not a league table. Side by side, though, they show how wide the range is.

The US publishes some of the most detailed national surveillance. Across 16 monitoring sites in 2022, the median age at the earliest known autism diagnosis was 47 months. It ranged from 36 months at the California site to 69.5 months at the Laredo, Texas site: nearly three years apart within one national network. Only 50.3% of autistic 8-year-olds with evaluation records had been evaluated by 36 months of age^[11].

Progress has been slow. In 2023, the CDC noted that “the reported median age of identification has not changed much over the years of ADDM Network surveillance”, though that does not necessarily mean early identification efforts have failed^[17]. The UK database study found no decrease between 2004 and 2014^[14]. There are signs of change. In the US, children born in 2018 had 1.7 times the cumulative incidence of an autism diagnosis or special-education eligibility by 48 months as children born in 2014 (22.6 compared with 13.1 per 1,000)^[11]. In Denmark, age at diagnosis moved younger in more recent birth cohorts^[15].

Some lower figures come from samples that are young by design, such as Australia's under-7s or Shenzhen's children already in rehabilitation services (even there, about one in six were not diagnosed until preschool age or later)^[16]. The English figures include adults, whose story we cover in diagnosed in adulthood.

Girls, children without intellectual disability and families with fewer resources wait longer

Girls. In English GP records, the median age at first recorded diagnosis was 8.64 years for males and 10.82 years for females^[4]. In Denmark, new diagnoses in boys peaked at age 5. In girls, rates stayed low until about age 8 and then rose steadily to a peak at 14^[15]. Canada's 2019 survey, though, found no statistically significant difference by sex^[12]. Our article on autistic women and girls looks at why girls are so often recognised late.

Children without intellectual disability. In the US, autistic children with intellectual disability were diagnosed at a median of 43 months, compared with 49 months for those without^[11]. A study of 969 caregivers in Pennsylvania found an average age at diagnosis of 3.1 years for autistic disorder but 7.2 years for Asperger's disorder. Children with significant language delays were diagnosed 1.2 years earlier on average^[18]. In Shenzhen, later diagnosis was more likely for children without intellectual impairment and for those whose autistic traits were rated as less pronounced^[16]. The children recognised latest appear to be those who fit a narrow picture of autism least well.

Ethnicity and migration. Among 406 children in Philadelphia's Medicaid system who received services in 1999, white children were diagnosed at 6.3 years on average and Black children at 7.9 years^[19]. But the direction of the gap varies by country. In Canada, children from visible minority groups were diagnosed younger, at a median of 2.7 years compared with 4.4 years^[12]. In Australia, children from culturally and linguistically diverse backgrounds were diagnosed 5 months earlier, with no difference between Indigenous and non-Indigenous children^[13]. In Shenzhen, children from migrant families had more than 9 times the odds of a delayed diagnosis compared with children holding a local household registration^[16]. Our piece on who gets identified goes further.

Income and health coverage. In Pennsylvania, children in near-poor families were diagnosed 0.9 years later than those with incomes more than double the poverty line, and rural children 0.4 years later than urban children^[18]. The 1990–2012 review linked higher socioeconomic status with earlier diagnosis^[5]. In Latin America, children with public rather than private health coverage were diagnosed later^[6]. Yet in the UK database study, lower socioeconomic status went with earlier diagnosis^[14], and Canada found no significant difference by household income^[12]. So the income gap does not look inevitable; it seems to depend on how easily families without money, time or connections can navigate the system.

Recognition could come years earlier

CDC guidance for clinicians states that “by age 2 years, a diagnosis by an experienced professional can be considered very reliable”^[20]. In a US study of 1,269 toddlers referred from the general population through universal screening in primary care or community referral, autism diagnoses had an overall stability of 0.84 and became stable from 14 months of age^[21]. The Australian researchers sum it up: “There may be a substantial gap between the age at which a reliable and accurate diagnosis of ASD is possible and the average age that children are currently diagnosed”^[13].

That gap matters because a diagnosis is so often what unlocks understanding, adjustments and services. The WHO states that “timely access to early evidence-based psychosocial interventions can improve the ability of autistic children to communicate effectively and interact socially”^[2]. The process also weighs on families. Just over half the UK parents surveyed were dissatisfied with it, and the time taken to reach a diagnosis was one of the factors linked to their satisfaction^[9].

Timing is also linked to mental health. In the UK Millennium Cohort Study, 396 of 11,320 14-year-olds had an autism diagnosis, and around nine in ten of them had cognitive ability in the typical range. Compared with non-autistic peers, those diagnosed after age 11 had 3.58 times the odds of depression at 14 (on a self-report questionnaire). Those diagnosed between 7 and 11 had 2.21 times the odds. For those diagnosed by age 5, the estimate was not statistically significant. Those diagnosed after 11 also had 3.16 times the odds of having self-harmed in the past year, the strongest association in the study^[22]. The study is observational and cannot prove cause, though the authors' analysis suggested reverse causation was unlikely to explain the link. They conclude that the results “suggest the importance of earlier diagnosis in preventing secondary mental health problems in this population”^[22]. Being recognised, understood and supported makes a real difference, and help is available now, whatever age someone was diagnosed.

What the evidence asks of us

Parents notice early; systems respond late; and the delay falls hardest on children who do not fit a narrow picture of autism and families who find services hardest to reach. The evidence asks services, funders and policymakers to:

• Make developmental monitoring routine. The WHO recommends “the monitoring of child development as part of routine maternal and child health care”^[2]. CDC guidance says all children should be screened specifically for autism at their 18- and 24-month well-child visits^[20].
• Start the clock at parents' first concern. Reviewers call for “interventions aimed at streamlining the process from first concern to eventual diagnosis”^[5]. European researchers call for shorter delays in screening and diagnosis, and for autism training for first-line professionals such as health visitors, GPs and paediatricians^[8].
• Look for the children most often missed. Girls, children without intellectual disability, minority and migrant communities, rural and low-income families. UK researchers conclude that “effective clinical strategies are needed to identify children with characteristics that have in the past delayed ASD diagnosis”^[14], and the earlier review calls for “strategies that target underserved populations”^[5].
• Count it. Few countries publish age at diagnosis, or the time from first concern to diagnosis, broken down by sex, ethnicity, income and region. What is not measured is hard to fix.
• Don't make support wait for a diagnosis. Where a diagnosis is the key to support, every month of delay is a month of support withheld.

None of this is about making children fit a template. It is about making sure autistic children, and the families who first noticed, are understood and supported years sooner. More evidence is gathered in our Autism & Health research library.

Language: this piece uses identity-first language (“autistic people”, “autistic children”), the preference of most autistic-led organisations. It uses clinical terms such as “autism spectrum disorder”, “autistic disorder” or “Asperger's disorder” only when naming a study or the diagnostic category it used.

Why we're publishing this

Health Insurance UK is a commercial health-insurance resource, not a charity and not a campaign. We compiled this evidence because autistic people, their families and the organisations pushing for earlier, fairer recognition deserve the international numbers clean, sourced and in one place, ready for a commissioner, a minister or a journalist. The case here belongs to the community. Our part is to stand alongside it with the data, and to say plainly what the figures show: families see it early, and systems take years to catch up.

How to read this data

The figures come from very different sources: meta-analyses, US surveillance records, Danish registers, English GP records, a Canadian parent-reported survey, Australian programme registrations, and caregiver surveys and clinic samples. Some report means and some medians; a few very late diagnoses can pull a mean well above the median. Samples limited to young children, such as under-7s, children in early-support services or Canada's survey of 1–17-year-olds, will show younger ages. This is partly because children who will be diagnosed later have not yet been counted. Samples that include adults, such as the English GP data, will show older ages. So the table is not a ranking, and figures from different years and methods are not a trend. Age at first concern is usually recalled by parents, sometimes years later. The 2022 income review pooled studies from several decades, and its income comparison rests on just 17 and 13 studies. So “no significant difference” there means a difference could not be shown, not that none exists. The UK mental-health figures are odds ratios compared with non-autistic peers, from an observational cohort. Evidence from low- and middle-income countries is thin and comes mostly from small clinic samples.

Use this data

Free to cite with attribution to Health Insurance UK. Charities, campaigners and journalists are welcome to reuse these figures and charts. 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. Matos MB, Bara TS, Cordeiro ML. Autism spectrum disorder diagnoses: a comparison of countries with different income levels. Clinical Epidemiology, 2022. europepmc.org/article/MED/35992506
2. World Health Organization. Autism (fact sheet). WHO, 2025. who.int/news-room/fact-sheets/detail/autism-spectrum-disorders
3. van 't Hof M, Tisseur C, van Berckelear-Onnes I, et al. Age at autism spectrum disorder diagnosis: a systematic review and meta-analysis from 2012 to 2019. Autism, 2021. europepmc.org/article/MED/33213190
4. 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, 2023. discovery.ucl.ac.uk/10168057
5. Daniels AM, Mandell DS. Explaining differences in age at autism spectrum disorder diagnosis: a critical review. Autism, 2014. europepmc.org/article/MED/23787411
6. Montiel-Nava C, Montenegro MC, Ramirez AC, et al. Age of autism diagnosis in Latin American and Caribbean countries. Autism, 2024. europepmc.org/article/MED/36602228
7. Bello-Mojeed MA, Omigbodun OO, Bakare MO, Adewuya AO. Pattern of impairments and late diagnosis of autism spectrum disorder among a sub-Saharan African clinical population of children in Nigeria. Global Mental Health, 2017. europepmc.org/article/MED/28596906
8. Mendez MA, Oakley B, Canitano R, et al. Autism care pathway in Europe. European Psychiatry, 2023. europepmc.org/article/MED/37694810
9. Crane L, Chester JW, Goddard L, Henry LA, Hill E. Experiences of autism diagnosis: a survey of over 1000 parents in the United Kingdom. Autism, 2016. europepmc.org/article/MED/25810370
10. Constantino JN, Abbacchi AM, Saulnier C, et al. Timing of the diagnosis of autism in African American children. Pediatrics, 2020. europepmc.org/article/MED/32839243
11. Shaw KA, 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, 2025. cdc.gov/mmwr/volumes/74/ss/ss7402a1.htm
12. Public Health Agency of Canada. Autism spectrum disorder: highlights from the 2019 Canadian Health Survey on Children and Youth. Government of Canada, 2022 (current PDF edition dated 2025). canada.ca/…/autism-spectrum-disorder-canadian-health-survey-children-youth-2019.pdf
13. Bent CA, Dissanayake C, Barbaro J. Mapping the diagnosis of autism spectrum disorders in children aged under 7 years in Australia, 2010–2012. Medical Journal of Australia, 2015. europepmc.org/article/MED/25832158
14. Brett D, Warnell F, McConachie H, Parr JR. Factors affecting age at ASD diagnosis in UK: no evidence that diagnosis age has decreased between 2004 and 2014. Journal of Autism and Developmental Disorders, 2016. europepmc.org/article/MED/27032954
15. Bliddal M, Gram EB, Sonne H, et al. Trends and incidence rates of neurodevelopmental disorders in Danish children and adolescents 2000–2024. European Child & Adolescent Psychiatry, 2026. europepmc.org/article/MED/41665685
16. Leng LL, Zhu YW, Zhou LG. Explaining differences in autism detection timing: age of diagnosis and associated individual and socio-familial factors in Chinese children. Autism, 2024. europepmc.org/article/MED/37491952
17. Maenner MJ, et al. Prevalence and characteristics of autism spectrum disorder among children aged 8 years — Autism and Developmental Disabilities Monitoring Network, 11 sites, United States, 2020. MMWR Surveillance Summaries, 2023. cdc.gov/mmwr/volumes/72/ss/ss7202a1.htm
18. Mandell DS, Novak MM, Zubritsky CD. Factors associated with age of diagnosis among children with autism spectrum disorders. Pediatrics, 2005. europepmc.org/article/MED/16322174
19. Mandell DS, Listerud J, Levy SE, Pinto-Martin JA. Race differences in the age at diagnosis among Medicaid-eligible children with autism. Journal of the American Academy of Child and Adolescent Psychiatry, 2002. europepmc.org/article/MED/12447031
20. Centers for Disease Control and Prevention. Clinical screening for autism spectrum disorder. CDC, 2025. cdc.gov/autism/hcp/diagnosis/screening.html
21. Pierce K, Gazestani VH, Bacon E, et al. Evaluation of the diagnostic stability of the early autism spectrum disorder phenotype in the general population starting at 12 months. JAMA Pediatrics, 2019. europepmc.org/article/MED/31034004
22. Hosozawa M, Sacker A, Cable N. Timing of diagnosis, depression and self-harm in adolescents with autism spectrum disorder. Autism, 2021. europepmc.org/article/MED/32772703