The Cost of Autism Support: Millions a Lifetime, Mostly Not Healthcare

Economists estimate that supporting an autistic person over a lifetime costs between about £0.9 million and £1.5 million in the UK, and similar sums in the US. Look closely and most of that money is not healthcare. It is special education, housing and support in adulthood, and earnings lost by autistic people shut out of work and by parents who cut their hours to fill gaps in services.

These are not figures about what autistic people cost. They are figures about what it costs when support comes late, comes in the wrong form, or never comes. It is the parent who gives up a job because no school place fits. It is the autistic adult who wants to work but never gets past an interview built for someone else. It is the family paying privately for help while a public service keeps them waiting. Read that way, the big totals are less a bill than a map of where systems fail, and where investment could change what happens next.

Lifetime estimates of support run to millions per person

The most widely quoted estimate comes from a 2014 study in JAMA Pediatrics. Combining data on prevalence, intellectual disability and living arrangements with the costs of services and lost earnings, it put the lifetime cost of supporting an autistic person with an intellectual disability at $2.4 million in the US and £1.5 million in the UK. For an autistic person without an intellectual disability, the figures were $1.4 million and £0.92 million^[1]. Assuming 40% of autistic people have an intellectual disability, and leaving out benefit payments, the UK total came to £3.1 billion a year for children and £29 billion a year for adults, at 2011 prices. The US equivalents were $61 billion and $175 billion^[1].

An earlier UK study by some of the same team put discounted lifetime costs of support at about £1.23 million with an intellectual disability and £0.80 million without^[4]. A 2007 US study put the additional lifetime societal cost associated with autism at $3.2 million per person, in 2003 dollars^[5]. A 2024 national study in China reported discounted lifetime costs of $2.65 million without an intellectual disability and $4.61 million with one, at 2020 prices, and a national total of $41.8 billion in 2020^[6]. Those figures are given in US dollars; the published summary does not say how yuan were converted, and they should not be read against the UK and US estimates as if they were like for like. In Australia, a 2011 study for the Autism Early Intervention Outcomes Unit put direct and indirect costs, before adding a value for reduced quality of life, at A$4.2 billion to A$7.3 billion a year in December 2010 dollars, depending on how many people were assumed to be autistic^[7]. Different methods, price years and currencies mean none of these estimates is directly comparable with another; the reasons are set out below.

The UK figures are old: a 2024 report from the London School of Economics (LSE) still quotes the 2014 total of more than £32 billion a year, at 2011 prices^[8]. We found no newer UK lifetime estimate.

Most of the cost is not healthcare

What a study counts matters more than its headline. In the 2014 UK–US study, the biggest costs for children were special education and parents' lost earnings. In adulthood they were residential care or supported living, and the autistic person's own lost earnings^[1]. In the UK, services made up 56% of the total, lost employment for the autistic person 42% and caregiver time 2%. In the US the split was 79%, 12% and 9%^[1].

A 2007 US estimate is starker. Medical care, including behavioural therapies, made up just 9.7% of the discounted lifetime cost. Lost earnings made up 59.3%: 30.7% the autistic person's own and 28.6% their parents'. Non-medical care and support made up 31%, with adult care alone at 21%^[5].

Using US national survey data on children, one team linked autism to $3,020 a year in extra healthcare costs but $14,061 in extra costs outside healthcare, including $8,610 in school costs (2011 dollars); 76% of autistic children used special education services, against 7% of other children. The authors concluded that studies focused on healthcare had underestimated the costs, especially for schools^[2].

In the Australian estimate, healthcare, social services and education together made up around 12% of the total, which also put a money value on reduced quality of life. The biggest tangible costs were lower incomes from reduced employment and unpaid care by family and friends^[7]. In China, lost productivity drove costs for autistic people without an intellectual disability. For those with one, the main costs were non-medical, such as rehabilitation and adult care^[6].

None of this means healthcare doesn't matter; autistic people have real and often unmet health needs. In a Minnesota birth cohort followed to age 38, medical costs were higher for autistic children and teenagers in every group studied. In adulthood, once co-occurring conditions were taken into account, the difference stayed significant in only 12% of groups^[9]. Those conditions are health needs that deserve good care.

Why the estimates vary so much

A review of around 50 cost studies from the US, UK, Australia, Canada, Sweden, the Netherlands and elsewhere found education a major cost for families^[10]. Which costs a study includes, and how it values them, explains much of the spread:

• Base year. The main lifetime estimates are priced in 2003 dollars^[5], 2011 pounds and dollars^[1] and 2020 dollars^[6].
• Discounting. Lifetime figures shrink future costs to today's values. The US study used 3% a year^[5] and the UK–US study 3.5%^[1]. The higher the rate, the lower the lifetime total.
• What is counted. Unpaid family care can dwarf everything else: in one European model it made up almost 90% of the six-year societal cost for an autistic child receiving usual care in England (€443,552 of €506,403, 2020 prices)^[11]. The Australian study added A$3.9 billion a year as a value for reduced quality of life^[7]. The UK–US study left out benefit payments, which move money rather than use it up^[1].
• Intellectual disability. Lifetime estimates for autistic people with an intellectual disability were 60 to 70% higher than for those without^[1]. So the share assumed to have one, 40% in the UK–US national totals, drives the overall figure. See our article on autism and intellectual disability.
• Lifespan and prevalence. The UK–US study assumed a life expectancy of 67 years^[1]. National totals also move with the prevalence assumed: the Australian range reflected 36.9 to 62.5 per 10,000^[7]. One US forecast put annual costs in 2025 at $461 billion, within a range of $276 billion to $1,011 billion^[12].
• Extra or total cost. Some studies count only costs over and above those for non-autistic people^[5][7]; the Chinese study describes a "gross" cost-of-illness approach, though its summary also refers to incremental lifetime costs, so what its headline figures represent is not clear from the abstract^[6].

After a change of method raised their estimate, the Australian authors cautioned "against making direct comparisons" with their earlier figures^[7]. The same caution applies to the table below.

ID: intellectual disability. Methods, years and currencies differ: not directly comparable.

Much of the cost is the price of unmet need

In US data from 2002–2008, mothers of autistic children earned 35% less than mothers of children with another health limitation, and 56% less than mothers of children with none. Family earnings were 28% lower than in families of children without health limitations. There were no significant differences for fathers^[13]. A 2024 study comparing the US with Norway found US mothers of autistic children were 12 to 25 percentage points less likely to be in work than mothers of children without special healthcare needs, depending on how their child's autism was rated. In Norway, the reported gap was 13 points, for mothers of children whose autism the study rated as "moderate/severe". The authors credited Norway's welfare state and policy package for the smaller gap^[14]. How much income families lose depends on policy.

In Western Australia, the median family cost was A$34,900 a year, and almost 90% of it (A$29,200) was lost income from work^[15]. The LSE estimates that reduced employment costs carers of young autistic children around £5,800 a year each. For parents of autistic children under six, that adds up to about £560 million^[8]. In Ontario, a 2025 pilot survey found families' money problems often begin during the diagnostic process, made worse by inadequate public support^[16]. Our article on the health of parents and carers covers the wider toll.

Education is one of the largest costs of childhood^[1][2]. In England, autism is now the most common primary need among pupils with an education, health and care (EHC) plan: one in three (33.5%) in 2025/26^[17].

In adulthood, lost earnings are among the largest costs^[1]. Only around three in ten autistic people in the UK are in work, yet three-quarters of those out of work say they want a job^[3]. Cost studies count this as a cost of autism; it is just as much a measure of how far hiring and workplaces shut autistic people out. See the autism employment gap.

Timing matters too. In a small Dutch study of 36 children, annual costs fell from €6,513 before diagnosis to €5,060 after, though the change was not statistically significant, and healthcare costs halved^[18]. The LSE team found no economic evidence at all on diagnostic delay, stressing that this "does not imply a lack of need"^[8].

Investing earlier and better can pay for itself, the models suggest

Does meeting need earlier pay? There is less evidence than there should be, and most of it comes from models rather than long-term trials, but what exists mostly points one way.

Early communication support. The Preschool Autism Communication Trial (PACT) tested a parent-mediated, communication-focused therapy with 152 preschool children. At 13 months, service costs were £4,489 higher per child than for usual care. The gap in wider societal costs was smaller (£1,385) and not significant, because families gave less unpaid care; on those results PACT was not shown to be cost-effective^[19]. A later simulation model extended the analysis to six years across England, Ireland, Italy and Spain and projected that PACT would save money from a societal point of view, mainly through less unpaid parental care^[11]. The LSE converts that modelled saving to £43,050 per person at 2024 prices^[8]. These six-year figures are projections, not observed savings: the resource-use data covered only the first 13 months^[11].

Support into work. A UK model found supported employment for autistic adults gave better outcomes than day services for an extra £18 per additional week in work, or £5,600 per quality-adjusted life year. Once potential savings on accommodation, NHS and social services were included, it gave better outcomes at lower total cost^[20]. For comparison, NICE usually recommends treatments costing less than £30,000 per quality-adjusted life year^[8]. Pro Bono Economics estimates that doubling the autistic employment rate from 30% would bring about 100,000 more autistic people into work and £900 million to £1.5 billion a year in economic benefits to society. The lower figure allows for up to 40,000 non-autistic people missing out on jobs^[3].

Specialist support for adults. In 2009, the National Audit Office modelled specialist teams for autistic adults without a learning disability in England. If such teams identified and supported about four per cent or more of this group locally, they could become cost-neutral across public spending over time; the model gave a better than 80% chance of savings, and more than 99% at an eight per cent identification rate^[21]. The LSE calls this evidence "quite old"; a more recent evaluation found one-to-one mental health support from specialist teams was cost-effective^[8].

Not every good service will save money, and it does not have to. As the LSE authors put it, a service can be worth funding even without savings "if the outcome gains are sufficient to justify the higher costs". They also judged that most studies probably underestimate the benefits, because they are short and some miss effects on families and work^[8].

What the evidence asks of us

Read carefully, cost studies do not show that autism is "expensive". They show that a great deal of money is already being spent and, as the LSE puts it, "it remains unclear whether this spending delivers the best possible outcomes" for autistic people and their families^[8].

• Talk about the cost of support and the cost of unmet need, not the cost of autistic people. How much is spent, and on what, is a policy choice.
• Budget across government departments. Costs spread across so many service systems that they raise "questions about coordination of services and sectors"^[1]. The gains from supported employment land in several budgets at once: lower benefit spending and higher tax receipts, and potential savings on housing, NHS and social services^[3][20]. Spending reviews should count both.
• Scale up what has been shown to work. The LSE report urges the UK government to use the National Autism Strategy to prioritise autism in funding, and to scale up services with an economic case, including PACT, positive parenting, Individual Placement and Support for employment, and mindfulness-based therapy^[8].
• Count families. Parents' lost earnings and unpaid care are among the largest costs, and the US–Norway comparison suggests policy can shrink them^[14].
• Fund better evidence, shaped by autistic people. Autism-specific economic evaluations are needed across every sector, shaped by autistic people and families^[8], starting with the cost of diagnostic delay.

More evidence is collected in 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" only when naming a study or diagnostic category.

Why we're publishing this

Health Insurance UK is a commercial health-insurance resource, not a charity and not a campaign. Cost-of-autism figures are often quoted without context, sometimes in ways that cast autistic people as a problem to be managed. We pulled the main studies together so that autistic people, families and the organisations working with them can see what the numbers are made of, and use the evidence that investing earlier and better improves lives and, on the evidence so far, can pay for itself. The case belongs to the community; our job is to put the evidence behind it.

How to read this data

Cost-of-illness studies estimate the total or extra costs linked to autism; economic evaluations compare the costs and outcomes of a particular service. The lifetime estimates are models, mostly at 2003–2011 prices, built on assumptions about prevalence, intellectual disability and lifespan that may no longer hold: treat them as orders of magnitude. The evidence comes mainly from the US, UK and Australia, with single studies from China, Canada, Norway, the Netherlands, Ireland, Italy and Spain. Several economic evaluations are small or modelled beyond their data, as noted.

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, financial or legal 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. Buescher AVS, Cidav Z, Knapp M, Mandell DS. Costs of autism spectrum disorders in the United Kingdom and the United States. JAMA Pediatrics, 2014. jamanetwork.com/journals/jamapediatrics/fullarticle/1879723
2. Lavelle TA, Weinstein MC, Newhouse JP, et al. Economic burden of childhood autism spectrum disorders. Pediatrics, 2014. pmc.ncbi.nlm.nih.gov/articles/PMC7034397
3. Gomez R, Sheikh S. Opening opportunities: Improving employment prospects for autistic people. Pro Bono Economics, for Autistica, 2023. pbe.co.uk/publications/opening-opportunities-improving-employment-prospects-for-autistic-people
4. Knapp M, Romeo R, Beecham J. Economic cost of autism in the UK. Autism, 2009. europepmc.org/article/MED/19369391
5. Ganz ML. The lifetime distribution of the incremental societal costs of autism. Archives of Pediatrics & Adolescent Medicine, 2007. jamanetwork.com/journals/jamapediatrics/fullarticle/570087
6. Zhao Y, Lu F, Wang X, et al. The economic burden of autism spectrum disorder with and without intellectual disability in China: a nationwide cost-of-illness study. Asian Journal of Psychiatry, 2024. europepmc.org/article/MED/38176313
7. Synergies Economic Consulting. Economic Costs of Autism Spectrum Disorder in Australia: Updated Study. Prepared for the Autism Early Intervention Outcomes Unit, 2011. synergies.com.au/wp-content/uploads/2019/09/AEIOU-Economic-Costs-of-Autism-Spectrum-1.pdf
8. Knapp M, Cyhlarova E, Salehi N, et al. The Economic Case for Prioritising Autism in Policy and Reform. Care Policy and Evaluation Centre, London School of Economics and Political Science, 2024. lse.ac.uk/cpec/assets/documents/Autismeconomics.pdf
9. Leibson C, Weaver A, Myers S, et al. Objective estimates of direct-medical costs among persons aged 3 to 38 years with and without research-defined autism spectrum disorder ascertained during childhood: a population-based birth-cohort study. Value in Health, 2020. europepmc.org/article/MED/32389225
10. Rogge N, Janssen J. The economic costs of autism spectrum disorder: a literature review. Journal of Autism and Developmental Disorders, 2019. europepmc.org/article/MED/30976961
11. Tinelli M, Roddy A, Knapp M, et al. Economic analysis of early intervention for autistic children: findings from four case studies in England, Ireland, Italy, and Spain. European Psychiatry, 2023. europepmc.org/article/PMC/PMC10594363
12. Leigh JP, Du J. Brief report: Forecasting the economic burden of autism in 2015 and 2025 in the United States. Journal of Autism and Developmental Disorders, 2015. europepmc.org/article/MED/26183723
13. Cidav Z, Marcus SC, Mandell DS. Implications of childhood autism for parental employment and earnings. Pediatrics, 2012. europepmc.org/article/MED/22430453
14. Brekke I, Alecu A, Ohrazda C, Lee J. Implications of childhood autism spectrum disorder for maternal employment: United States vs. Norway. Maternal and Child Health Journal, 2024. europepmc.org/article/MED/38864988
15. Horlin C, Falkmer M, Parsons R, et al. The cost of autism spectrum disorders. PLOS ONE, 2014. europepmc.org/article/MED/25191755
16. Xu J, Graziosi G, Rodrigues FF, et al. Economic impacts of caring for autistic children in Ontario, Canada: report from a pilot study. Frontiers in Public Health, 2025. europepmc.org/article/MED/41041377
17. Department for Education. Special educational needs in England, academic year 2025/26. Explore education statistics, 2026. explore-education-statistics.service.gov.uk/find-statistics/special-educational-needs-in-england
18. ten Hoopen LW, de Nijs PFA, Greaves-Lord K, et al. Quality of life and societal cost in autistic children: an exploratory comparative study pre- and post-diagnosis. Journal of Autism and Developmental Disorders, 2026. europepmc.org/article/MED/40057931
19. Byford S, Cary M, Barrett B, et al. Cost-effectiveness analysis of a communication-focused therapy for pre-school children with autism: results from a randomised controlled trial. BMC Psychiatry, 2015. europepmc.org/article/MED/26691535
20. Mavranezouli I, Megnin-Viggars O, Cheema N, et al. The cost-effectiveness of supported employment for adults with autism in the United Kingdom. Autism, 2014. europepmc.org/article/MED/24126866
21. National Audit Office. Supporting people with autism through adulthood (report and technical paper). NAO, 2009. nao.org.uk/reports/supporting-people-with-autism-through-adulthood