Getting diagnosed as an adult

What a defensible assessment contains

The diagnostic criteria come from the American Psychiatric Association’s DSM-5-TR. The assessment process comes from the main adult-ADHD clinical guidelines outside the US — NICE NG87 (National Institute for Health and Care Excellence) and CADDRA 4.1 (Canadian ADHD Resource Alliance). Both call for a clinical interview with a developmental history, information from someone who knew you as a child where possible, a mental health screen for comorbidities, and a medical review to rule out conditions that mimic ADHD; both treat rating scales as a supplement, not a verdict. A structured interview and a written report are this site’s standard; neither guideline requires them. There is no US adult ADHD guideline yet: APSARD’s (American Professional Society of ADHD and Related Disorders) is still in development (CHADD). Five or more of nine inattentive and/or hyperactive-impulsive symptoms at the adult threshold, several present before age 12, in at least two settings, with functional impairment, not better explained by another disorder. Expect 60–180 minutes total, often split across two sessions.

A defensible assessment produces a written report — diagnosis (or differential with reasoning), ICD-10/11 or DSM-5-TR codes, functional impairment documentation, comorbidity findings, treatment recommendations, accommodations recommendations where relevant. The report is what the reader will need for Americans with Disabilities Act (ADA) accommodations, insurance authorisation, disability claims, or future re-verification. A clinician who does not produce one is leaving the reader without the operational output of the assessment.

DIVA-5, CAARS, ASRS — what each is for

DIVA-5 (Diagnostic Interview for Adult ADHD, version 5) is the most widely used structured diagnostic interview for adult ADHD globally — developed by J.J. Sandra Kooij and the Dutch DIVA Foundation, semi-structured, clinician-administered, aligned to DSM-5 (Kooij et al. 2019 European consensus⁸). The patient-facing version is preparation, not diagnosis. Free for clinical use in many jurisdictions; standard of care in the Netherlands, much of the UK private sector, increasingly in US adult ADHD specialty practices.

A 2026 study of 217 adults first diagnosed in adulthood (J Clin Med 2026)³ compared self- and parent-ratings on the DIVA-5 and found self-ratings consistently higher than parent-ratings across all symptom and impairment domains. Anxiety levels notably moderated the gap — especially for childhood and adult hyperactivity-impulsivity. Practical read: bringing an informant who knew you young matters more when current anxiety is high, because the assessor will be reading both readings against your anxiety load.

CAARS (Conners’ Adult ADHD Rating Scales) and ASRS (Adult ADHD Self-Report Scale; Kessler et al. 2005⁶) are screens, not diagnoses. Neither is diagnostic on its own; both are adjuncts. The clinician using a rating scale as the diagnosis is doing it wrong.

In-person vs telehealth — protocol over modality

Modality is not the determining variable; protocol and clinician are. Cunningham 2024¹⁴ and other concordance studies show good agreement between in-person and synchronous-video adult ADHD evaluation when evaluator credentials and protocol are equivalent. A 2.5-hour video evaluation by a psychiatrist using DIVA-5 with collateral interview is operationally equivalent to the same evaluation in person.

Platform models are a separate question. The 15–30 minute video intake by a rotating nurse practitioner under per-script compensation is not the modality those concordance studies validated. The regulatory record lives at telehealth ADHD diagnosis. The same questions apply to any clinician regardless of modality: credential, expected duration, structured interview used, collateral handling, comorbidity screening, written report.

What to bring

• Childhood records where available. Report cards (elementary teacher comments — “doesn’t apply herself,” “easily distracted”), paediatrician notes, any educational assessments. Many late-diagnosed adults do not have these; absence is not disqualifying: NICE NG87 advises against excluding ADHD on the basis of an uncertain time of onset.
• A collateral informant. Parent (ideal for childhood history), older sibling, long-term partner, close friend. 15–30 minute call. Increasingly clinicians accept written statements where a call is not possible.
• Family psychiatric and medical history. ADHD, autism, anxiety, depression, bipolar, substance use in first-degree relatives; cardiovascular events. ADHD heritability is ~74% (Faraone & Larsson 2019¹³) — family history is meaningful evidence.
• A symptom timeline and three to four specific recent examples. The missed deadline, the forgotten appointment, the half-finished project, the emotional reaction to feedback. Having them ready prevents the “I can’t think of an example right now” freeze.
• Questions for the clinician. Cost, follow-up structure, who manages medication if applicable, whether they produce a written report, second-opinion practices.

How long it takes

Pre-appointment forms: 45–90 minutes of the reader’s own time, days before. Initial evaluation: 60–180 minutes total, often split across two appointments — first session covers history, structured interview, comorbidity screening; second covers feedback, diagnostic formulation, treatment planning. Collateral interview: 15–30 minutes, scheduled separately. Medical mimic workup (thyroid-stimulating hormone (TSH), ferritin, B12, basic metabolic panel; sleep study if indicated): 1–2 weeks for labs, 2–8 weeks for sleep study where access is limited. Feedback and report delivered 1–3 weeks after the final session.

Total elapsed booking-to-report: 2–8 weeks if the clinician has availability; substantially longer with waitlists. US specialty waits often 2–6 months; UK National Health Service (NHS) 1–4+ years since ~2020; NHS Right to Choose pathway 4–16 weeks via approved private providers; Canada CADDRA-network 3–12 months.

The cost picture

The variability is structural. US cash-pay psychiatrist evaluation $300–$2,500 depending on geography and credential; psychologist $400–$3,000; comprehensive neuropsychological battery $1,500–$5,000 — NICE NG87 bases diagnosis on a clinical and psychosocial assessment, developmental and psychiatric history, and observer reports, and lists no neuropsychological battery; it is worth paying for where executive function (EF) or learning-disability questions are present. Insurance coverage varies by plan, by state, and by network status. UK private assessment £500–£1,200; NHS no cost but long wait. Two state-level US patterns worth flagging: California has generally broader Medicaid coverage of behavioural-health evaluation; Texas, Florida, and Pennsylvania have narrower coverage and longer specialty waits.

If they say no

A “no” from a properly conducted assessment is a meaningful clinical finding, not by default a clinician error. The differential for adult ADHD includes generalised anxiety, major depression, bipolar II, sleep disorders, thyroid dysfunction, iron deficiency, trauma sequelae, autism without ADHD, and substance use. A clinician who has worked through that differential and landed elsewhere has done the job.

The right of second opinion is real and uncontroversial, particularly where you feel the assessment did not cover your presentation adequately. The meaningful question to ask before pursuing one: did the assessment contain the components above? If the developmental history was skipped, collateral was not pursued, no validated rating scale was used, comorbidity was not screened, medical mimics were not ruled out — the answer was procedurally weak and a second opinion is appropriate. If all components were present and the clinician concluded otherwise, the answer is more likely accurate than not.

The age-of-onset debate

DSM-5 lowered the age-of-onset threshold from before age 7 (DSM-IV) to before age 12. Moffitt et al. 2015¹¹ and Caye et al. 2016¹² found adults meeting criteria in birth cohorts without childhood ADHD, sparking the “late-onset ADHD” debate. Sibley et al. 2018¹⁰ followed a clinical cohort from ages 10 to 25 and concluded that late-onset cases may be valid, but more commonly reflect nonimpairing cognitive fluctuations, a comorbid disorder, or substance use. NICE NG87 says exclusion based on an uncertain time of onset is not recommended, so the practical position is to assess impairment carefully while still seeking developmental evidence. The article does not adjudicate; the reader should know the debate exists when clinicians refer to it.