ADHDAutism

More Diagnoses Are Not the Same as Overdiagnosis: A Clinical Response to the Independent Review

Listen to the Article

0:00 / 0:00

The 2026 Independent Review cannot conclude ADHD is overdiagnosed. A consultant psychiatrist holds it to its caveats on impairment, delay and assessment.

By Dr Hossein Rostamipour, Consultant Psychiatrist · Eton Psychiatrists

Why we have written this. The Independent Review into prevalence and support for mental health conditions, ADHD and autism, chaired by Professor Peter Fonagy with Professor Gillian Baird and Professor Sir Simon Wessely, was published on 9 October 2026. Its recommendations will shape how ADHD and autism assessment is commissioned in England for years. This is a consultant psychiatrist's response, written for patients, families, referrers and commissioners who will read the headlines rather than the report. It is general information, not personal medical advice.

Watch and explore

A short animated video summary of this article and an interactive version are in production and will appear here.

The Independent Review into prevalence and support for mental health conditions, ADHD and autism is a careful document, and better than its coverage. It says it cannot conclude that overdiagnosis is occurring at scale, that the evidence does not show people presenting today are less impaired than before, and that people seeking help are not the problem.

I write as a consultant psychiatrist in an independent clinic assessing adults with ADHD under NHS Right to Choose. That interest is declared, and I have seen provision in my own sector that fell short of what patients deserved. My purpose is to hold the review to its own caveats, because the commissioning that follows will read its recommendations, not its qualifications.

Why diagnoses rose

Administrative prevalence of ADHD in 12 to 17 year olds rose from 0.26 per cent to 3.88 per cent between 2000 and 2025, against a population estimate the review puts, with appropriate caution, at around 3 to 4 per cent. Diagnosis has been catching up with need, not outrunning it.

DSM-IV barred co-diagnosis with autism and framed the condition around the hyperactive boy; DSM-5 lifted both constraints. Women with the inattentive presentation, often treated for years for anxiety or depression, began to be assessed, with growing recognition that perimenopause can unmask long-compensated symptoms. Older adults, assumed to have grown out of a childhood condition, were seen too. A system that once found a tenth of its ADHD population was not being prudent; it was missing most of them. Diagnosed prevalence will not rise indefinitely, and the national prevalence survey the review recommends is the right way to know when it has stopped.

Impairment and the threshold for assessment

The review proposes prioritising assessment by need, functional impairment and risk, and says people with significant impairment should be assessed within a reasonable time. I agree, with two cautions.

First, NICE guideline NG87 sets referral at moderate impairment, and a prioritisation framework will read moderate as deferrable. Much of the impairment I assess arises where stable traits meet environments built around sustained attention and self-organisation: classrooms, open-plan offices, the benefits system. Adults hold down jobs and relationships through compensatory effort the review itself acknowledges. Masking is impairment paid for privately, and the bill arrives as burnout, anxiety or depression in general mental health services. Waiting for collapse before assessing is not prioritisation.

Second, prioritising by acuity has a limit the review does not address: the most distressed are the least assessable. An ADHD diagnosis rests on a six-month pattern of symptoms traced to childhood. A person in crisis or intense emotional dysregulation cannot give a valid account of that pattern, and the clinician cannot separate trait from state. That person needs stabilising support first, and stimulant titration belongs in a period of stability. Triage should aim to stabilise and then assess, never to divert.

What delay costs

ADHD is unusual in psychiatry in having a treatment with a large, rapid and well-replicated effect: 70 to 90 per cent of patients respond to stimulant medication. Swedish registry data associate starting medication with 19 per cent lower all-cause and 25 per cent lower unnatural-cause mortality in the two years that follow (Li et al., 2024); later work from the same group shows lower rates of suicidal behaviour, substance misuse, transport accidents and criminality (Zhang et al., 2025). The review finds childhood ADHD carries a fourfold risk of being not in education, employment or training, and that 44 per cent of young people who are NEET now have a work-limiting condition.

NHS England's ADHD Taskforce puts the cost of untreated ADHD to the UK at around £17 billion a year and finds that only 15 per cent of adults with ADHD receive medication. Coaching and psychological strategies have a place, but the NG87 evidence review found too little evidence to recommend them as stand-alone treatment, and most of my patients had exhausted self-help long before referral. With more than 800,000 open ADHD and autism referrals in England, and seven in ten people deteriorating after six months of waiting, not assessing is not a saving. It is a transfer of cost to crisis teams, GPs, employers and the Department for Work and Pensions.

Needs-led, with the diagnosis

"Needs-led plus diagnosis" is sound if the second half is honoured. The review says support must not be withheld while people wait and that no one should leave a list without being connected to help. I would add that support without assessment is generic, and generic support offered to people impaired in planning, organisation and memory carries its own attrition: each extra screen and re-justification loses some of them. A good assessment is needs-led in itself. The person who does not meet criteria leaves with a formulation and a direction, and the person who does leaves with a treatment that works.

The independent sector

Around 28 providers deliver adult ADHD assessment under Right to Choose, according to ADHD UK's provider table. None operates without CQC registration, which is a legal precondition of holding an integrated care board contract, and each works under an NHS Standard Contract to NG87 at prices and pathways the ICB sets. Ratings are mixed, as they are across the NHS, and I support consistent regulation, including the government's stated intention to bring every ADHD and autism diagnostic assessment under CQC regulation.

The review's sharper point is that misdiagnosis is likelier when assessment is brief, not face to face, fragmented or focused on a single condition. Most Right to Choose assessment is by video, including ours, and the challenge is fair. The answer is not the medium but what is done within it: adequate time, structured instruments, developmental and informant history, screening for autism and for the mood, anxiety and trauma presentations that mimic ADHD, and one clinician from assessment into treatment. These are auditable, and they should be audited.

Conclusion

The review should be judged by whether people are seen sooner, assessed well and treated. Read as licence to divert, it will produce a longer route to the same door, and a public narrative about overdiagnosis that stigmatises people the review itself says are no less impaired than before. I would rather help it be read the other way.

How we approach this at Eton Psychiatrists

We are a CQC-registered, consultant-led service assessing ADHD and autism together, across the lifespan, under NHS Right to Choose in England and under our contract with NHS Wales, alongside private practice. Every assessment takes the time the review asks for, uses structured instruments and developmental and informant history, screens for the conditions that mimic ADHD, and is followed by treatment from the same clinician. Children and adolescents are seen in London and in Manchester at Alderley Edge, as well as online.

Where to go next

Book an appointment: https://etonpsychiatrists.co.uk/book-appointment
The ADHD boom: overdiagnosed, or finally understood? https://etonpsychiatrists.co.uk/blog/the-adhd-boom-overdiagnosed-or-finally-understood
What unrecognised ADHD costs the UK: https://etonpsychiatrists.co.uk/blog/cost-of-unrecognised-adhd-uk
Video or in person? Choosing the assessment that lets you be seen: https://etonpsychiatrists.co.uk/blog/adhd-autism-assessment-video-or-in-person
Adult ADHD assessment and treatment: https://etonpsychiatrists.co.uk/services/adhd-adult
Right to Choose: https://etonpsychiatrists.co.uk/services/right-to-choose

Summary

The review is a careful document, and better than its coverage. It says it cannot conclude that overdiagnosis is occurring at scale, that people presenting today are no less impaired than before, and that people seeking help are not the problem. My purpose is to hold it to its own caveats, because the commissioning that follows will read its recommendations, not its qualifications. Diagnoses rose because a system that once found a tenth of its ADHD population started finding more of them. Masking is impairment paid for privately. The most distressed are the least assessable, so triage should stabilise and then assess, never divert. Delay has a cost that is paid by crisis teams, GPs, employers and the DWP. And the right response to variable quality in the independent sector is to audit what is done inside an assessment, not to dictate the medium.

Sources

Department of Health and Social Care. Independent review into prevalence and support for mental health conditions, ADHD and autism: final report (Fonagy, P., Baird, G. & Wessely, S.). 9 October 2026. https://www.gov.uk/government/publications/independent-review-into-prevalence-and-support-for-mental-health-conditions-adhd-and-autism
Li, L., Zhu, N., Zhang, L. et al. ADHD pharmacotherapy and mortality in individuals with ADHD. JAMA 331(10), 850–860 (2024). https://doi.org/10.1001/jama.2024.0851
National Institute for Health and Care Excellence. Attention deficit hyperactivity disorder: diagnosis and management. NICE guideline NG87 (2018, updated 2019). https://www.nice.org.uk/guidance/ng87
NHS England. Report of the independent ADHD Taskforce: Part 1 (April 2025) and Part 2 (November 2025). https://www.england.nhs.uk/long-read/report-of-the-independent-adhd-taskforce-part-1/
Zhang, L., Zhu, N., Sjölander, A. et al. ADHD drug treatment and risk of suicidal behaviours, substance misuse, accidental injuries, transport accidents, and criminality: emulation of target trials. BMJ 390, e083658 (2025). https://doi.org/10.1136/bmj-2024-083658
ADHD UK. Right to Choose: provider table. https://adhduk.co.uk/right-to-choose/ (accessed September 2026).
The ideas, clinical content and text of this article are the authors' own, drawn from their clinical practice. AI tools were used only to help structure and proofread it, under the authors' direction, and every claim was checked by the authors before publication.

Frequently Asked Questions

Did the Independent Review find that ADHD is overdiagnosed?

No. The review says it cannot conclude that overdiagnosis is occurring at scale, that people presenting today are no less impaired than those who presented before, and that people seeking help are not the problem. It does warn that a focus on diagnosis rather than need could shift the balance towards misdiagnosis if current trends continue, which is an argument for better assessment rather than less of it.

Why have ADHD diagnoses risen so much?

Because the diagnostic rules changed in 2013 to allow ADHD and autism to be diagnosed together and to recognise the inattentive presentation, and because awareness caught up a decade later. Women, older adults and inattentive children who had been missed began to be assessed. Diagnosed prevalence in adolescents rose from 0.26 per cent in 2000 to 3.88 per cent in 2025, which is still at or below the review's own estimate of how common ADHD is.

Will I be deprioritised if my impairment is "moderate"?

NICE sets referral at moderate impairment, and the review proposes prioritising by need and risk. The concern raised here is that moderate will be read as deferrable, when much moderate impairment is masked effort that eventually presents as burnout, anxiety or depression. If that describes you, say so plainly at referral.

I am in crisis. Should I have an ADHD assessment now?

Usually not immediately. A valid assessment needs a reliable account of a lifelong pattern, which is hard to give in a crisis, and stimulant titration belongs in a period of stability. The right sequence is stabilising support first, then assessment. That is different from being diverted away from assessment altogether.

Is a video assessment less reliable than a face-to-face one?

The review notes that misdiagnosis is likelier when assessment is brief, not face to face, fragmented or single-condition. The medium is the least important of those four. What matters is adequate time, structured instruments, developmental and informant history, screening for mimics and co-occurring autism, and continuity with one clinician. Those can be delivered and audited by video or in the room, and your own preference should decide which.

Where can I get help?

Through your GP, who can refer to NHS services or, under Right to Choose in England, to a provider like us; or privately. If you are in crisis, call 999 or go to A&E; for urgent help call NHS 111, or Samaritans free on 116 123.