Oct 1, 2026·Special Population & Related Conditions
How Is Autism Assessed in Adults? The Adult Diagnostic Pathway
An adult autism assessment weighs DSM-5-TR criteria against a clinical interview, structured observation and whatever developmental history can be recovered.
Dr. Russell T. WarneChief Scientist
Share
An adult autism assessment is a clinical judgement built from a diagnostic interview about present functioning, a structured observation of social behaviour, whatever developmental history can still be reconstructed, and a worked differential against conditions that produce a similar picture in adulthood. A questionnaire score decides who joins the queue for it, and the published evidence says those questionnaires sort people badly.
The childhood pathway, which has its own page on this site, is a different exercise. There the developmental history is current, the caregiver is in the room, and school records are a phone call away. An adult arriving at 40 may have no living informant, no surviving school file, and three decades of learned strategies sitting on top of the behaviour a clinician needs to see. Everything distinctive about adult assessment follows from that.
What the DSM-5-TR requires, and why Criterion C is the hard part
Autism spectrum disorder in the DSM-5-TR is one diagnosis with severity levels, and an adult must meet the same criteria a child does: deficits in social communication, plus restricted or repetitive patterns of behaviour, interests or activities, with severity rated separately for each domain.
Criterion C is where the adult case gets difficult. The manual requires that symptoms "must be present in the early developmental period," then adds that they "may not become fully manifest until social demands exceed limited capacities, or may be masked by learned strategies in later life." That clause is the DSM conceding, inside the criterion itself, that the evidence a clinician needs may have been papered over for twenty years.
Two points apply only to adults. A patient holding a DSM-IV diagnosis of Asperger's disorder or PDD-NOS carries forward to autism spectrum disorder under the DSM-5-TR, so an old label does not need re-litigating. And measured ability spans the full range within this diagnosis, as our article on IQ and autism sets out; cognitive testing here documents functioning rather than establishing the diagnosis.
The screeners do not triage the way their manuals promise
The AQ with its 10-item short form, and the Ritvo Autism Asperger Diagnostic Scale-Revised, are the self-report instruments an adult is most likely to be handed before a referral. Both looked excellent when first published and both have performed badly on the people who turn up in clinic.
• The developer figures: At a cut-point of 6 the adult AQ-10 showed sensitivity of 0.88 and specificity of 0.91, with cases drawn from a research centre's volunteer database. The RAADS-R was reported at 97% sensitivity and 100% specificity above a threshold of 65.
• The AQ in clinic: Across 476 adults seen consecutively at a national specialist service in London, 73% of whom were diagnosed, the AQ-10 at a cut-off of 6 gave sensitivity 0.77 (95% CI 0.72 to 0.82) against specificity of 0.28 (0.20 to 0.38). It did not predict diagnosis better than chance, and nearly two-thirds of those scoring below the cut-off were diagnosed anyway.
• The RAADS-R in clinic: In 50 service users of an NHS specialist autism service it produced 100% sensitivity against 3.03% specificity, a positive predictive value of 34.7%, and an area under the curve of 0.45 (95% CI 0.285 to 0.612), indistinguishable from guessing.
• A head-to-head: Tested together in 210 referred patients and 63 controls, the RAADS-R and two short AQ forms all ran "much lower than the values reported in the literature," and none had sufficient validity to predict a diagnosis in outpatient settings.
A high score tells a service you belong in the queue. A low score, among people who referred themselves because something has never fitted, clears almost nobody.
Structured observation when there is no childhood informant
Two instruments carry the diagnostic weight, and only one of them survives the absence of a parent.
• ADOS-2 Module 4: The module for verbally fluent adolescents and adults runs a conversation and activity sequence under graded social pressure, and our guide to the ADOS-2 describes the module structure. Module 4 shipped without the revised algorithm and calibrated severity scores the other modules received; a 2014 revision supplied them, with sensitivity and specificity exceeding 80%. An independent study of 93 adult men found the original algorithm at the standard cut-off of 7 gave sensitivity of 0.61 with specificity of 0.82, separated autism from psychopathy and typical development, and lost specificity against schizophrenia with marked negative symptoms.
• The ADI-R: The caregiver interview is scored against behaviour around ages 4 and 5, so it needs someone who was there, and our page on the ADI-R sets out the algorithm. In that London service it was performed first whenever the parent or caregiver "was alive and willing to be interviewed" and the patient consented. It was completed for 305 of 476 patients, or 64%.
When no parent is available a competent assessment substitutes what exists: school reports, an older sibling or aunt, childhood medical notes, dated recollections. None of it reaches the standard of an administered ADI-R, and a good report says so.
Camouflaging, and the delay in identifying women
Camouflaging covers the strategies people use to conceal autistic characteristics or to pass socially. The Camouflaging Autistic Traits Questionnaire, built from autistic adults' own accounts and administered to 354 autistic and 478 non-autistic adults, resolved into 25 items across three factors: Compensation, Masking and Assimilation. Internal consistency was 0.94 and test-retest reliability 0.77. The CAT-Q measures a behaviour and is not a diagnostic instrument.
The consequence shows up in age at diagnosis. In pooled online samples of 242 autistic men and 570 autistic women aged 18 to 75, women were diagnosed later on average, with a significant camouflaging-by-sex interaction in which high-camouflaging women were diagnosed later still. A Dutch survey of 2,275 people found that among respondents aged 18 and over, women with autistic disorder had been identified later than men, and a systematic review of barriers for girls and young women named compensatory behaviour and clinician bias among the recurring obstacles.
For a clinician this changes the interview rather than the criteria. Asking what a social situation costs, how it is rehearsed beforehand and what the recovery afterwards looks like recovers what an observation of competent surface behaviour misses.
The differential, and the co-occurrence that is not a differential
Several conditions dominate the adult differential, and each can also be present at once, which is the part handled badly.
• ADHD: A meta-analysis of 96 studies put pooled ADHD prevalence in the autism population at 28% (95% CI 25 to 32). DSM-IV forbade diagnosing ADHD alongside a pervasive developmental disorder and the DSM-5 removed that exclusion, so the question is which conditions are present rather than which one wins.
• Social anxiety: The same meta-analysis put anxiety disorders at 20%, depressive disorders at 12% and obsessive-compulsive disorder at 9%. The discriminating question is developmental: social anxiety centres on fear of negative evaluation and usually has a datable onset, where autistic social difficulty predates the anxiety and persists in situations the patient is not worried about.
• Personality disorders: Among 122 consecutively referred adults of normal intelligence with an autism spectrum diagnosis, 62% met lifetime criteria for at least one DSM-IV personality disorder and 35% for two or more, most often obsessive-compulsive at 32%, avoidant at 25% and schizoid at 21%. That is a specialist clinic under a retired diagnostic system, so read it as evidence of large overlap rather than as population rates.
Schizophrenia spectrum disorders sat at 4%, and the ADOS specificity problem against negative symptoms is a reminder that the hardest differentials are those where observable behaviour genuinely converges.
Who may diagnose, how long the wait is, and what the diagnosis unlocks
In the United States there is no federal credential for this. Diagnosis is made by licensed clinical psychologists, psychiatrists, neurologists and some developmental specialists, and scope of practice is set by state licensure statute, so who can diagnose you depends on where you live. England specifies the structure centrally, with NICE recommending a specialist autism team and a corroborative informant account where one can be obtained.
The capacity gap is documented in the negative. A CDC-authored modelling study opens by stating that national and state estimates of adults living with autism "are nonexistent due to the lack of existing surveillance systems funded to address this need," then estimates 2.21% of US adults aged 18 to 84, diagnosed and undiagnosed. A review of adult services concluded that clinical provision for autistic adults is markedly limited, so long queues follow and self-identification fills the space they leave.
Self-identification and a formal diagnosis do different work. A study of 665 adults diagnosed or self-diagnosed in adulthood found that fear of not being believed by professionals was the most frequent and most severe reported barrier. What the formal document adds is leverage, and the machinery for an adult is employment law rather than special education. Under the ADA regulations a reasonable accommodation is a modification to the job application process, to the work environment, or to the manner in which a position is customarily performed, and an employer's obligation is triggered by a request supported by documentation of a disability.
Frequently asked questions
Can an online questionnaire diagnose adult autism?
No. The AQ-10 and the RAADS-R are referral filters whose specificity in clinic samples has run as low as 0.28 and 3%. A score is a reason to seek an assessment.
What if my parents cannot give a developmental history?
That is common enough that a national specialist service completed an ADI-R for only 64% of its adult referrals. A clinician should work from school records, other relatives and childhood medical notes, and should say in the report what was unavailable.
Can you be diagnosed with both autism and ADHD?
Yes. DSM-IV prohibited it and the DSM-5 removed the exclusion. Pooled ADHD prevalence in the autism population is about 28%.
Why are women identified later?
Camouflaging accounts for part of it. Among autistic adults aged 18 to 75, women were diagnosed later than men, and high-camouflaging women later again.
Is self-identification valid?
It is a reasonable working hypothesis and many are later confirmed, but it does not create the footing that documented disability gives you when you request a workplace accommodation.
The takeaway
The adult pathway is the childhood pathway with its best evidence removed. The observation instrument is weaker at this age, the interview carrying the developmental history often cannot be administered at all, the screeners deciding who gets seen have specificity near zero in referred populations, and years of camouflaging stand between the clinician and the behaviour. A good assessment answers that by being explicit about its own gaps. Ask which ADOS-2 module was used, whether a childhood informant was available, what alternative explanations were ruled out, and which parts of the history rest on your recall alone. To see how a carefully normed cognitive measure reports its results, you can take an online IQ test built by psychometricians.
References
1. American Psychiatric Association. (2022). Diagnostic and statistical manual of mental disorders (5th ed., text rev.). American Psychiatric Association Publishing. doi.org
2. Allison, C., Auyeung, B., & Baron-Cohen, S. (2012). Toward brief "red flags" for autism screening: The short Autism Spectrum Quotient and the short Quantitative Checklist in 1,000 cases and 3,000 controls. Journal of the American Academy of Child and Adolescent Psychiatry, 51(2), 202-212. doi.org
3. Ritvo, R. A., Ritvo, E. R., Guthrie, D., Ritvo, M. J., Hufnagel, D. H., McMahon, W., Tonge, B., Mataix-Cols, D., Jassi, A., Attwood, T., & Eloff, J. (2011). The Ritvo Autism Asperger Diagnostic Scale-Revised (RAADS-R): A scale to assist the diagnosis of autism spectrum disorder in adults. Journal of Autism and Developmental Disorders, 41(8), 1076-1089. doi.org
4. Ashwood, K. L., Gillan, N., Horder, J., Hayward, H., Woodhouse, E., McEwen, F. S., Findon, J., Eklund, H., Spain, D., Wilson, C. E., Cadman, T., Young, S., Stoencheva, V., Murphy, C. M., Robertson, D., Charman, T., Bolton, P., Glaser, K., Asherson, P., Simonoff, E., & Murphy, D. G. (2016). Predicting the diagnosis of autism in adults using the Autism-Spectrum Quotient (AQ) questionnaire. Psychological Medicine, 46(12), 2595-2604. doi.org
5. Jones, S. L., Johnson, M., Alty, B., & Adamou, M. (2021). The effectiveness of RAADS-R as a screening tool for adult ASD populations. Autism Research and Treatment, 2021, 9974791. doi.org
6. Sizoo, B. B., Horwitz, E. H., Teunisse, J. P., Kan, C. C., Vissers, C., Forceville, E., Van Voorst, A., & Geurts, H. M. (2015). Predictive validity of self-report questionnaires in the assessment of autism spectrum disorders in adults. Autism, 19(7), 842-849. doi.org
7. Hus, V., & Lord, C. (2014). The Autism Diagnostic Observation Schedule, Module 4: Revised algorithm and standardized severity scores. Journal of Autism and Developmental Disorders, 44(8), 1996-2012. doi.org
8. Bastiaansen, J. A., Meffert, H., Hein, S., Huizinga, P., Ketelaars, C., Pijnenborg, M., Bartels, A., Minderaa, R., Keysers, C., & de Bildt, A. (2011). Diagnosing autism spectrum disorders in adults: The use of Autism Diagnostic Observation Schedule (ADOS) module 4. Journal of Autism and Developmental Disorders, 41(9), 1256-1266. doi.org
9. Hull, L., Mandy, W., Lai, M.-C., Baron-Cohen, S., Allison, C., Smith, P., & Petrides, K. V. (2019). Development and validation of the Camouflaging Autistic Traits Questionnaire (CAT-Q). Journal of Autism and Developmental Disorders, 49(3), 819-833. doi.org
10. Milner, V., Colvert, E., Hull, L., Cook, J., Ali, D., Mandy, W., & Happé, F. (2024). Does camouflaging predict age at autism diagnosis? A comparison of autistic men and women. Autism Research, 17(3), 626-636. doi.org
11. Begeer, S., Mandell, D., Wijnker-Holmes, B., Venderbosch, S., Rem, D., Stekelenburg, F., & Koot, H. M. (2013). Sex differences in the timing of identification among children and adults with autism spectrum disorders. Journal of Autism and Developmental Disorders, 43(5), 1151-1156. doi.org
12. Lockwood Estrin, G., Milner, V., Spain, D., Happé, F., & Colvert, E. (2021). Barriers to autism spectrum disorder diagnosis for young women and girls: A systematic review. Review Journal of Autism and Developmental Disorders, 8(4), 454-470. doi.org
13. Lai, M.-C., Kassee, C., Besney, R., Bonato, S., Hull, L., Mandy, W., Szatmari, P., & Ameis, S. H. (2019). Prevalence of co-occurring mental health diagnoses in the autism population: A systematic review and meta-analysis. The Lancet Psychiatry, 6(10), 819-829. doi.org
14. Hofvander, B., Delorme, R., Chaste, P., Nydén, A., Wentz, E., Ståhlberg, O., Herbrecht, E., Stopin, A., Anckarsäter, H., Gillberg, C., Råstam, M., & Leboyer, M. (2009). Psychiatric and psychosocial problems in adults with normal-intelligence autism spectrum disorders. BMC Psychiatry, 9, 35. doi.org
15. Lewis, L. F. (2017). A mixed methods study of barriers to formal diagnosis of autism spectrum disorder in adults. Journal of Autism and Developmental Disorders, 47(8), 2410-2424. doi.org
16. Dietz, P. M., Rose, C. E., McArthur, D., & Maenner, M. (2020). National and state estimates of adults with autism spectrum disorder. Journal of Autism and Developmental Disorders, 50(12), 4258-4266. doi.org
17. Murphy, C. M., Wilson, C. E., Robertson, D. M., Ecker, C., Daly, E. M., Hammond, N., Galanopoulos, A., Dud, I., Murphy, D. G., & McAlonan, G. M. (2016). Autism spectrum disorder in adults: Diagnosis, management, and health services development. Neuropsychiatric Disease and Treatment, 12, 1669-1686. doi.org
18. National Institute for Health and Care Excellence. (2021). Autism spectrum disorder in adults: Diagnosis and management (Clinical guideline CG142). nice.org.uk
19. Equal Employment Opportunity Commission. (2024). Regulations to implement the equal employment provisions of the Americans with Disabilities Act, 29 CFR 1630.2 (definition of reasonable accommodation at 1630.2(o)). Electronic Code of Federal Regulations. ecfr.gov
Hero image: an empty doctor's office waiting room, by Kurt Kaiser, released under CC0 1.0 (creativecommons.org/publicdomain/zero/1.0/deed.en). Via Wikimedia Commons.
Take our professional IQ test
Want to know your IQ? Try the first ever professional online IQ test.
