UK autism data

At a glance

Useful current figures, kept in context

Each item names its nation, period and source. Figures shown together are not necessarily comparable.

Autism assessment and diagnosis

England has a recurring suspected-autism referral series

NHS England’s latest published data, for June 2026, show 294,792 people with an open referral for suspected autism. Of these, 256,017 had been open for at least 13 weeks. These are referral-pathway figures, not counts of people diagnosed with autism.

The series is useful but needs careful reading. Its open-referral method changed from the September 2023 publication. A provider began flowing suspected-autism referrals in December 2025, creating a clear break in the national trend. The figures should not be used as a simple measure of diagnosis prevalence or as a UK waiting-list total.

NHS England — Autism Statistics, July 2025 to June 2026

Where comparable national figures are not available

ZenEmu has not made a four-nation referral comparison. Wales’ available children’s neurodevelopmental information combines ADHD and autism and is management information rather than official statistics. Scotland does not have a suitable recurring national autism-referral series; CAMHS waiting-time data exclude neurodevelopmental pathway activity. Northern Ireland’s quarterly referral and positive-diagnosis series is currently paused and incomplete following the Encompass rollout, with missing Trust data in the latest release.

Welsh Government — neurodevelopmental assessment information · Public Health Scotland — CAMHS waiting times · Northern Ireland Department of Health — Quarterly Autism Statistics

Children and education

Education systems record different things

England records a primary special educational need. Scotland records reasons for additional support, and a pupil may have more than one. Northern Ireland’s annual school-age series records diagnosed autism. Wales records ALN/SEN need in maintained schools while its system is moving from SEN to ALN. These figures are useful within their own systems; they should not be added together or treated as a four-nation prevalence table.

Exclusions and attainment

DfE’s 2024/25 exclusions release records 42,677 suspension events and 395 permanent exclusions involving pupils with autistic spectrum disorder as their primary SEN need. A suspension is an event, not a unique pupil, so this should not be read as a percentage of all autistic pupils.

Department for Education — Suspensions and permanent exclusions in England, 2024/25

ZenEmu monitors England’s attainment-by-primary-need tables, but has not made attainment a headline figure in this initial page. The most useful comparison needs careful contextual treatment, not a bare scorecard.

Benefits and social security

Administrative classifications are not prevalence data

Benefit statistics describe claims and payments within a particular system. They do not count everyone with an autism diagnosis, and a recorded primary condition does not show every condition a person has.

Employment

No current recurring autism-specific employment measure selected

ZenEmu monitors official disabled-employment statistics, but has not found a current recurring official series that gives a sufficiently robust autism-specific employment estimate to place alongside administrative figures. Older sample-survey tables are not a substitute for a maintained measure, and would need their sample-size and confidence-interval context to be meaningful.

This is an intentional gap, not a claim that employment data are unimportant.

Health and inpatient data

Hospital data describe a much narrower population

At the end of July 2026, the Assuring Transformation dataset recorded 1,215 people in the autism category in mental health hospitals in England. The same official dataset separately records people in the learning-disability category and the learning-disability-and-autism category. ZenEmu does not relabel the pooled population as “autistic people”.

NHS England — Learning Disability and Autism inpatient statistics, July 2026

ZenEmu also monitors LeDeR annual reporting. It is not presented as a general autism mortality database: it covers deaths notified to LeDeR, has a specific inclusion history and involves small numbers for autism-only analysis.

Limits

What these figures cannot tell us

• PIP and disability-payment data describe claims and payments, not autism prevalence.
• School data describe children and young people recorded within education systems; they do not automatically describe diagnosis or all autistic pupils.
• Referral statistics describe demand on an assessment pathway, not the number of people diagnosed.
• Health-service data describe people using those services, not autistic people generally.
• Absence from an administrative dataset does not mean absence of autism.

Method

How ZenEmu uses these figures

ZenEmu collates existing official statistics; it is not the producer of those statistics. Every published figure is checked against the linked primary release and recorded in a local monitoring register with its definition, period, geographical scope, publication details, source-check date and comparability notes.

Figures are not combined unless their definitions genuinely permit it. Publishing organisations may revise historical figures or change methodology. ZenEmu checks the monitored sources periodically and changes this page’s “Page last updated” date only when data, methodology or substantive content changes.

If you spot an error or a newer official publication, please use the existing ZenEmu contact route and name the figure or source that needs checking.

Maintenance

Recent data updates

12 September 2026 — Initial page published after an audit of the monitored official series.