What assessments are used to evaluate students for Autism?
Autism assessment in the UK blends clinical tools with school-based observations, so support can start early, even before a formal diagnosis. According to NHS advice, families usually begin by speaking with a GP or school SENCO, who can trigger referral into a multidisciplinary assessment pathway (NHS). NICE guidance recommends combining standardised tools with developmental history and reports from education and health professionals (NICE CG128, NICE CG170).
Core tools used in clinical pathways
Clinical pathways rely on structured tools that guide assessment, planning, and intervention across multidisciplinary teams. These tools ensure consistency in care, support evidence-based decision-making, and enhance outcomes for individuals receiving treatment.
- ADOS-2 (Autism Diagnostic Observation Schedule): a structured, play-based observation across modules by age/language. It captures social communication and restricted/repetitive behaviours; results must be interpreted alongside other evidence (overview example).
- ADI-R (Autism Diagnostic Interview—Revised): a detailed, structured interview with parents/carers that explores early development, communication, and behaviour (NICE CG128).
- Complementary tools sometimes used to profile strengths/needs: SCQ (screening questionnaire), CARS-2, GARS-3 (rating scales). These can guide referrals but don’t replace a full clinical assessment (see the National Autistic Society’s overview of tools: NAS).
What schools use
Schools do not diagnose autism. They gather evidence via teacher observations, communication/sensory checklists, and short screeners such as the SCQ to inform SEN support and referrals under the SEND Code of Practice (DfE SEND Code of Practice). This helps justify requests for specialist assessment or, where needed, an EHC needs assessment.
How evidence is combined
NICE advises a multidisciplinary approach: clinicians integrate ADOS-2/ADI-R results with medical and developmental history, school reports, and direct observation across settings (NICE CG128, NICE CG170). NHS England’s national framework aims to improve pathway quality and timeliness (NHS England framework).
What recent research says
Recent studies highlight how clinical pathways continue to evolve with evidence-based updates, digital innovations, and improved evaluation frameworks.
- BMJ Paediatrics Open (2024) – A national survey reported variable access and delays to assessment across regions; ADOS-2 and ADI-R remain core tools and stronger clinical–education data-sharing was recommended. Result: system capacity is the main bottleneck, not tool validity.
- Nature (2020) – Comparative analyses suggested ADI-R better reflects symptom severity while ADOS may capture sex/gender differences more sensitively. Result: using both improves characterisation of presentations.
- Systematic Review (2025, ScienceDirect) – Reported high sensitivity/specificity for ADOS-2/ADI-R; school screeners are useful for referral but are insufficient for diagnosis. Result: supports NICE’s combined-evidence approach.
- Lancet EBioMedicine (2025) – Early AI-based behavioural tracking showed promising accuracy. Result: not yet ready for routine clinical use; requires external validation.
When to Seek Extra Support
If families notice challenges like communication delays, emotional distress, or sensory issues, they should seek support. Parents can discuss concerns with the SENCO, who can begin early interventions. If needed, a GP or NHS assessment can help, and services like Autism Detect can provide additional screening and advice, complementing NHS and local authority pathways.
Takeaway
For children and students in the UK, autism evaluation centres on ADOS-2 and ADI-R within a multidisciplinary pathway, supported by school observations and screening to inform referral and support. Evidence from recent studies backs the combined-tools approach, while schools use screening to guide SEN support, not diagnosis. If you’re concerned, speak to the SENCO and your GP to start the referral process.

