Failing the Autism Act
In 2009, England made a landmark decision. Through the Autism Act — the first disability-specific legislation of its kind — it recognised that autistic people were being failed by systems that did not see them. The response was to name that failure and require government to act. Recognition was not symbolic: it was the foundation of accountability.
Seventeen years on, that foundation appears less secure. The Act requires government to publish and review a national strategy to meet the needs of autistic adults, alongside statutory guidance for local authorities and NHS bodies to fulfil the strategy. The current Autism Strategy is due to expire in July 2026, yet no timetable for renewal has been set. Statutory guidance was never updated to reflect the extension of rights to autistic children and young people. Key national commitments were not completed or extended. At the same time, autism is largely absent from major reform programmes across education, health and employment, while policy discourse has increasingly framed autism in terms of backlog, cost and overdiagnosis.
Taken together, these developments suggest something more fundamental than delay or oversight. They point to a gradual shift away from autism as a distinct policy priority. As the House of Lords Autism Act Committee concluded in its report Time to deliver: the Autism Act and the new autism strategy1, previous strategies did not fail for lack of intent, but because of weak accountability, inconsistent implementation and a lack of cross-government ownership.
The current Autism Strategy is due to expire in July 2026 without a timetable for renewal. Statutory guidance was never updated to reflect the extension of rights to autistic children and young people. Key national commitments were not completed or extended. Autism is largely absent from major reform programmes across education, health and employment, while policy discourse has increasingly framed autism in terms of backlog, cost and overdiagnosis.
This raises critical questions: not only whether systems are working for autistic people, but whether they are still being designed with a clear understanding of autistic need at all. Without that, there is little reason to expect they will deliver in practice.
The SEND system is increasingly shaped by neurodiversity, but not designed for it
One system where this tension is particularly visible is the Special Educational Needs and Disabilities (SEND) system. Growth in Education, Health and Care (EHC) plans is often described as a general increase in need, but a significant proportion of this rise is driven by autism. Autistic pupils now account for over a third of EHC plans in England, up from around a quarter a decade ago.2 In effect, the SEND system is increasingly functioning as a neurodiversity system, but without being explicitly designed, resourced or structured as such.
This creates a structural tension at the heart of current reforms. While there is a clear ambition to improve mainstream inclusion and reduce reliance on statutory processes, the conditions required to achieve this are not yet in place. Nearly all school leaders report insufficient resources to meet SEND need,3,4 and 84% of autistic young people and families rate teachers’ autism knowledge as below average or non-existent.5
At the same time, proposals within the Schools White Paper6 to introduce nationally defined Specialist Provision Packages and reserve statutory support primarily for children with the “most complex” needs risk narrowing the group of pupils considered eligible for specialist provision. The reforms are built on the assumption that more SEND children can be supported within mainstream settings if earlier interventions, training and inclusion measures are strengthened.
For autistic pupils, however, this assumption requires careful scrutiny. Autism does not always present in ways that fit neatly within standardised definitions of complexity. More fundamentally, autistic need does not always sit on a simple spectrum from lower to higher need. A pupil may be academically able while simultaneously needing significant sensory, social or therapeutic support to access education effectively. Children may require different types and intensities of support across different domains of their lives, making need more multidimensional than many categorisation systems allow.
This has important implications for reforms that seek to reserve specialist provision primarily for children with the “most complex” needs. Many autistic pupils currently supported in specialist settings are precisely those whose needs risk being reclassified under a narrower definition of complexity. Not because their needs have diminished, but because standardised thresholds are unlikely to adequately capture the variability and context-dependence of autistic need in practice.
The central question is therefore not simply whether these pupils continue to receive support, but whether reforms correctly identify and accept the type and intensity of support they require. Features often associated with specialist provision – smaller classes, highly specialist staff, structured environments and integrated therapeutic support – are not necessarily an escalation of support. For some autistic pupils, they may represent the minimum conditions needed to access education successfully. The risk is that reforms are built on assumptions about which children need specialist provision that prove incorrect in practice, leaving some autistic pupils without access to the support, environments and expertise they need to achieve good outcomes.
In effect, the SEND system is increasingly functioning as a neurodiversity system, but without being explicitly designed, resourced or structured as such.
Evidence from our own provision illustrates what is at stake. The Rise School, run by Ambitious about Autism, supports autistic pupils who can access a broadly age-related curriculum but require specialist environments to do so. All pupils have EHC plans, and many arrive following unsuccessful placements in mainstream schools. Their needs are often misunderstood rather than absent: many have co-occurring conditions, experience high levels of anxiety, and require consistent, structured support to engage in learning.
Under emerging reform proposals, these are among the pupils most likely to be considered suitable for mainstream settings, despite clear evidence that such placements have not worked for them previously. This creates a tension between the intended aim of reform — to improve inclusion — and its likely effect in practice.
The Rise School’s model is built around elements that are difficult to replicate without deliberate investment: autism-specialist staff, consistent relationships with key adults, and integrated therapeutic support. Long-term tracking shows that this approach delivers strong outcomes. In 2024/25, 93% of leavers progressed into education, training or employment. This stands in marked contrast to wider national outcomes, where only 17.5% of autistic young people transition into sustained employment and 48% progress into sustained education.7
This is not an argument for a single model of provision. It is evidence, drawn from practice, of a broader principle: outcomes improve when environments are designed around autistic need. The question for reform is whether removing access to such environments for a group of pupils — on the basis that their needs are no longer deemed sufficiently complex — would represent progress, or risk repeating patterns of exclusion and disengagement.
When unmet need is reframed as overdiagnosis
Waiting lists for autism assessment have risen significantly in recent years. As of March 2026, more than 270,000 people were waiting an autism assessment in England, including over 163,000 children and young people. Nearly 90% had been waiting longer than the recommended 13-week standard, with many waiting years for assessment and support.8
Alongside this, a new narrative has gained traction: that rising numbers of autism diagnoses reflect overdiagnosis rather than unmet need. This framing has increasingly appeared in public and political discourse, including comments from the previous Health and Social Care Secretary raising concerns about potential overdiagnosis.9
Last year, the Government commissioned an independent review into the prevalence and diagnosis of autism, ADHD and mental health, which published interim findings in March 2026.10 The review presents a far more complex picture than public debate often suggests. It concludes that rising diagnosis does not necessarily indicate rising prevalence. Multiple factors are likely contributing simultaneously: greater public awareness, improved recognition among previously under-identified groups, increasing self-identification, and the growing role diagnosis plays in accessing support within education and public services.
The review also highlights rising levels of emotional distress, anxiety and loneliness among young people. Taken together, these findings suggest that growing demand for diagnosis cannot be understood in isolation from wider social pressures, unmet need and the way support systems are currently structured.
Research conducted by Ambitious about Autism reflects many of these findings.11 Parents and carers consistently reported seeking assessment primarily to protect their child’s mental health. They tell us the path to diagnosis is fraught with difficulties, delays and frustrations – not a situation anyone would wish to find themselves in. Often parents enter the assessment process after there has been a significant deterioration in their child’s mental health. Nearly a third of families reported paying for private assessments due to long delays, with some taking out loans to access assessment.
Yet much of the public discussion surrounding the review centred on the idea that children are somehow ‘incentivised’ to seek a diagnosis, conjuring up the idea that they are somehow seeking a shallow reward, when in reality the incentive in question is just having their basic needs met. At the same time, much of the current discourse around the autism assessment system seems to be placing blame on parents for seeking answers to the challenges their children face. This shifts the debate away from how systems respond to need and towards whether that need is legitimate in the first place.
This distinction matters. Language shapes public perceptions and taken out of context can have a negative impact on understanding and support for autistic people. Reframing rising diagnosis primarily as overdiagnosis shifts attention away from systemic gaps in provision and towards reducing or narrowing recognition itself, whether through changes to diagnostic pathways or alternative models of assessment rather than improving support.
There is also a deeper consequence. Diagnosis is not only a gateway to services; it is a mechanism through which need becomes visible within education, health and employment systems. Reducing its scope risks creating a more ambiguous system in which access to support becomes less predictable, less consistent and less accountable. The focus should not be on controlling demand by reducing recognition of need, but on ensuring that need is identified and responded to earlier, more consistently and more effectively.
Reframing rising diagnosis primarily as overdiagnosis shifts attention away from systemic gaps in provision and towards reducing or narrowing recognition itself, whether through changes to diagnostic pathways or alternative models of assessment rather than improving support.
Outcomes that the system cannot ignore
If autism were peripheral to the system, these questions might remain abstract. Outcomes data suggests otherwise. Across education, autistic pupils experience significantly higher rates of absence, exclusion and disengagement. In 2024–25, 29.3% of autistic pupils were persistently absent, compared with 17.5% of non-autistic pupils. Severe absence was almost three times higher among autistic pupils (6.2% compared with 2.2%).12
These experiences are not isolated. Autism is not a mental health condition, but autistic young people are more likely to experience mental health challenges, with up to 70% meeting criteria for a mental health condition13. For many, unmet need in education contributes directly to declining wellbeing, disengagement and reduced confidence during critical developmental years. Nearly three-quarters (71%) report experiencing some form of lost learning, with one in six describing stepping away from school as the best available option for their mental health.14
The consequences extend into adulthood. Autistic people have the lowest employment rate of any disability group, at around 31.4%.15 Just 17.5% of autistic young people transition into sustained employment.16 At the same time, in 2025 around 30.1% of young people whose primary health condition was recorded as learning difficulties or autism were not in education, employment or training (NEET), compared with 12.6% of young people without those conditions.17
These outcomes reflect systemic barriers rather than lack of aspiration. Most autistic young people want to work and have skills to contribute. Yet they face a system that is not designed with their needs in mind, from inaccessible recruitment processes to inconsistent workplace adjustments and fragmented support across departments.
This creates a tension at the heart of current policy. On the one hand, rising economic inactivity is increasingly framed as a problem to be addressed through employment reform and welfare policy. On the other hand, the specific barriers faced by autistic young people are not explicitly recognised within these same frameworks. The shift from autism-specific work, such as the Buckland Review,18 to broader strategies like Get Britain Working has not been accompanied by a clear articulation of how autistic people’s barriers will be addressed.
The risk is that outcomes are treated in isolation from their causes. Without upstream action – including inclusive education, effective mental health support, and structured pathways into work – downstream interventions are unlikely to succeed. In this context, there is a danger that policy focuses on reducing inactivity through tighter expectations or conditionality, rather than addressing the barriers that prevent participation in the first place.
The recent interim Young People and Work Review,19 led by Alan Milburn, reinforces this point. Rather than attributing rising youth inactivity to a lack of motivation among young people, the review argues that young people are being failed by systems that are not providing the support they need to realise their ambitions. Significantly, it identifies SEND as central to understanding rising NEET rates, finding that young people with SEND are 80% more likely to be NEET and that having an EHC plan is the strongest single predictor of becoming NEET between ages 17 and 19.
Autistic people have the lowest employment rate of any disability group, at around 31.4%. Just 17.5% of autistic young people transition into sustained employment. In 2025 around 30.1% of young people whose primary health condition was recorded as learning difficulties or autism were not in education, employment or training (NEET), compared with 12.6% of young people without those conditions.
These findings should inspire autism policy, but they have not. A third of those EHC plans are for a primary need of autism. Yet, while the review rightly highlights structural barriers facing young people, there is still no discussion of the specific challenges many autistic young people encounter. Autism continues to be absorbed into wider categories, while the distinct barriers affecting autistic young people remain insufficiently recognised within policy responses.
This matters not only for fairness, but for effectiveness. Autistic young people represent an under-recognised and under-supported part of the workforce. Evidence shows that with the right adjustments – structured pathways, supported internships, and employer confidence – outcomes improve significantly, benefiting both individuals and the wider economy. Research from the London School of Economics20 estimates that doubling autism employment rates could generate annual economic gains of between £900 million and £1.5 billion, while analysis by Pro Bono Economics21 suggests that each previously unemployed autistic adult supported into work generates societal benefits worth approximately £15,400 per year.
The question, therefore, is not whether autistic people have ambition. The question is whether systems are designed to recognise and unlock that potential. Bridging this gap requires more than general reform. It requires recognising autistic people not only as a group at risk of poor outcomes, but as a group whose participation is critical to any credible strategy to reduce youth economic inactivity.
The limits of standardisation in a system defined by complexity
A common response to rising demand is to seek greater consistency: clearer categories, standardised pathways, and more tightly defined thresholds for support. These approaches can improve administrative clarity and may help manage short-term pressures. But they also reflect an underlying assumption: that need can be neatly classified and addressed through uniform solutions.
Elements of the Government’s SEND reforms illustrate this tension. The proposed Specialist Provision Packages are intended to create greater consistency by defining nationally the support available to children deemed to have the most complex needs. While this may help reduce variation between areas, it also raises a broader question about whether highly individual and context-dependent needs can be effectively captured through standardised categories and pathways. Greater consistency in what is provided does not necessarily translate into better outcomes if the underlying system struggles to recognise the full range of needs it is intended to support.
Autism is characterised by variability, context-dependence and interaction across multiple domains, including communication, sensory processing and learning. Needs do not sit comfortably within discrete categories, and effective support often depends on flexibility, adaptation and specialist understanding.
Efforts to simplify this complexity are understandable. But there is a risk that, in seeking to make the system more manageable, reform again moves away from what makes it effective. This reflects a broader policy choice: whether to invest in understanding and respond to complexity, or to attempt to reduce it.
The proposed Specialist Provision Packages are intended to create greater consistency by defining nationally the support available to children deemed to have the most complex needs. While this may help reduce variation between areas, it also raises a broader question about whether highly individual and context-dependent needs can be effectively captured through standardised categories and pathways.
There is a parallel challenge in how costs are understood. Investment in early, tailored support is often viewed through the lens of short-term affordability, rather than long-term value. Yet the costs of unmet need – persistent absence, exclusion, mental health crises and economic inactivity – are substantial. Without upstream investment in education, mental health and transitions, systems incur greater costs downstream, including increased reliance on crisis services and long-term welfare support.
In this context, the question is not whether the system can afford to invest in autism-specific support, but whether it can afford not to. The alternative is already expensive. When needs go unmet, the costs reappear elsewhere: in school absence, mental health crises, unemployment. Evidence suggests that improving employment outcomes alone could generate annual economic gains of up to £1.5 billion.22 The real policy choice is therefore not between spending and saving, but between investing early in support or paying later for the consequences of its absence.
Conclusion: recognition as a condition of reform
England’s SEND reforms are taking place at a moment of significant change. The ambition to build a more inclusive and sustainable system is clear. But reform is not only about structure; it is about what – and who – the system is designed to serve.
The success of the SEND reforms will ultimately depend on whether each component of the proposed system can recognise and respond to autistic need in practice. The Government’s vision for mainstream inclusion and reduced reliance on statutory processes depends on multiple interconnected reforms functioning effectively together: National Inclusion Standards, workforce training, Individual Support Plans, Specialist Provision Packages, guidance and accountability frameworks. The system will therefore only be as effective as its weakest point. If any element fails to recognise and respond appropriately to autistic need, the wider ambition of reform may prove illusory. Autistic pupils could find themselves facing the same patterns of exclusion, lost learning and disengagement, but with fewer routes to specialist support and statutory protection when mainstream provision falls short.
The Autism Act established a principle that remains relevant: recognition is not symbolic; it is the foundation of accountability. Only when needs are explicitly recognised can systems be designed around them, resourced to meet them, and held accountable for outcomes. When recognition weakens, responsibility becomes diffuse, delivery becomes inconsistent at best, and poor outcomes become easier to explain away.
The experience of the current Autism Strategy demonstrates that recognition on paper is not enough. The challenge was never simply whether autism was named in policy, but whether that visibility meaningfully influenced how systems were designed and how success is measured. A future Autism Strategy must therefore do more than articulate principles or retrospectively assemble autism-related commitments from elsewhere in government, reflecting decisions that have already been made. It must provide a strategic framework that shapes the design of policy across education, health and employment, ensuring autistic people’s needs are considered from the outset. As the House of Lords Autism Act Committee concluded,23 the challenge is not whether the Autism Act remains the right framework, but whether government is willing to deliver on it in practice.
A future Autism Strategy must provide a strategic framework that shapes the design of policy across education, health and employment, ensuring autistic people’s needs are considered from the outset.
This matters because major decisions are being taken now and the Government has indicated that it is waiting for them before refreshing the Autism Strategy.24 The SEND reform, the Young People and Work Review, the prevalence review and wider health reforms will all shape outcomes for autistic people for years to come. If autistic need is not recognised within these reforms as they are developed, the risk is not simply that autism becomes less visible in policy or a sudden disappearance of support. It is something more fundamental. It is that systems are built on assumptions that do not adequately reflect autistic need. And once these systems are established, correcting those assumptions becomes significantly harder.
Ultimately, this is a question of whether England can build on the original ambition of the Autism Act: to recognise need, design systems accordingly, and hold those systems accountable for outcomes. Or whether it continues down a path where autism is increasingly absent from the design of major reforms. The difference between those paths will determine whether autistic people are supported to thrive through systems designed around their needs or continue to experience the consequences of systems that fail to recognise their needs, with significant human and economic costs for individuals, families and society.
1. House of Lords Select Committee on the Autism Act 2009 (2025), ‘Time to deliver: The Autism Act 2009 and the new autism strategy’.
2. Department for Education, ‘Special Educational Needs in England statistical publications’ (2015 and 2026 editions).
3. NFER. (2024), The ongoing impact of cost-of-living crisis on schools.
4. The Guardian (2026), ‘Lack of specialist staff hinders support for Send children, teacher survey finds’.
5. Ambitious about Autism. (March 2025)’ ‘When will we learn: Lost Learning report’.
6. Department for Education, ‘Every child achieving and thriving’, 23 February 2026.
7. Department for Education, 9 May 2024, Destinations of key stage 4 and 5 students: 2022.
8. NHS England Digital (2026), ‘Autism Statistics, April 2025 to March 2026’.
9. BBC (16 March 2025). ‘Mental health conditions are overdiagnosed, Streeting says’.
10. Department of Health and Social Care (2026), ‘Independent review into mental health conditions, ADHD and autism: interim report’.
11. Ambitious about Autism (2026), ‘A system at breaking point: Parent and carer experiences of autism assessment and diagnosis’.
12. Department for Education, March 2026, ‘Pupil absence in schools in England: 2024 to 2025’ For pupils with a primary need of autism in state-funded primary and secondary schools.
13. Simonoff, E., Pickles, A., Charman, T., Chandler, S., Loucas, T. and Baird, G. (2008), ‘Psychiatric disorders in children with autism spectrum disorders: prevalence, comorbidity, and associated factors in a population-derived sample’, Journal of the American Academy of Child and Adolescent Psychiatry, 47(8), 921-929.
14. Ambitious about Autism. (March 2025), ‘When will we learn: Lost Learning report’.
15. Office for National Statistics, ‘The employment of disabled people 2025’ – GOV.UK
16. Department for Education, 9 May 2024, ‘Destinations of key stage 4 and 5 students: 2022’.
17. Department for Education, NEET statistics, March 2026.
18. Department for Work and Pensions, ‘The Buckland Review of Autism Employment: report and recommendations’, 28 February 2024.
19. Department for Work and Pensions, 28 May 2026. Young people and work; interim report.
20. LSE, ‘The Economic Case for Prioritising Autism in Policy and Reform’, Published December 2024.
21. Pro Bono Economics, ‘Opening opportunities: Improving employment prospects for autistic people’, Published December 2023.
22. LSE, The Economic Case for Prioritising Autism in Policy and Reform Published December 2024
23. House of Lords Select Committee on the Autism Act 2009 (2025), ‘Time to deliver: The Autism Act 2009 and the new autism strategy’.
24. House of Lords debate, ‘New Autism Strategy (Autism Act 2009 Committee Report)’, 10 June 2026.
