Three thousand complaints, and a curve that tripled in four years
On Tuesday the Parliamentary and Health Service Ombudsman published “Improving access: ADHD and autism assessment and treatment commissioning” — an analysis drawing on around 3,000 complaints about ADHD and autism services in England over six years. The headline curve: complaints rose from 410 in 2021/22 to 1,257 in 2025/26, an increase of more than 200% in four years. The Ombudsman’s verdict is delivered in institutional language, but it is not ambiguous: demand is severely outstripping capacity, care is fragmented, waits are long, and where someone lives — not what they need — substantially determines what they get. The report calls it a postcode lottery, and the Ombudsman, Paula Sussex, said people should not have to fight their way through a confusing system, with many feeling forced to pay privately — which, she noted, is not what good public services should look like.
Readers of this corpus will recognise the shape immediately, because the Ombudsman has now said institutionally what the research has been saying piecemeal all year. Three weeks ago I covered the Cardiff mortality review: the later the system finds you, the steeper the risk — a nine-fold hazard for adult diagnosis against essentially baseline risk for childhood diagnosis, with every unsupported year compounding the comorbidities that carry the danger. That review priced the unfound years. This report audits the machine that leaves people unfound — and finds it failing not incidentally but structurally, in ways that have names, dates and responsible bodies.
One concrete example makes the postcode lottery mechanical rather than metaphorical. In January, Greater Manchester moved to gate ADHD assessments by needs-based criteria — access to an assessment now depends on demonstrating sufficient impact, in a way it doesn’t forty miles away. Whatever the local justification, the result is exactly what the Ombudsman describes: the same person, with the same presentation, gets an assessment in one integrated care board’s area and a refusal in another. That is not a waiting list problem. It is a definition-of-entitlement problem, being solved forty-two different ways by forty-two different boards.
The legal right that patients were not told they had
The finding with the sharpest edge sits in the middle of the report: patients were not given clear, accessible information about their legal Right to Choose a provider. Not a discretionary scheme — a legal right, under which a patient referred for an ADHD or autism assessment in England can choose any qualifying provider holding an NHS contract, including providers with dramatically shorter waits. The Ombudsman found people simply weren’t told, and is now calling on government to publish national guidance and clarify how the right operates.
I wrote about this mechanism at length in the Right to Choose con — and the con, then as now, was never that the right doesn’t exist. It’s that the system treats the right as a problem to be managed rather than an entitlement to be honoured: referrals steered away from it, GPs unaware or unwilling, commissioners quietly resenting the invoices it generates. The report’s challenges chapter states the tension openly — the right sits in direct friction with integrated care boards’ discretion over local commissioning, and nobody at national level has resolved which wins. (For anyone who needs the practical mechanics rather than the politics, the Directory’s guide to using Right to Choose covers the referral wording and the qualifying providers.)
The shared-care corollary compounds it. People who do use the right, get assessed, and are titrated onto medication by an independent provider then hit a second wall: GPs declining shared-care agreements — sometimes on principle, sometimes on workload, sometimes on advice from the local board — leaving patients with a diagnosis, a working prescription, and no NHS route to continue it. The report documents this as a systemic pattern, not an anecdote. A right of access that terminates in a prescription nobody will take over is not access; it is a longer corridor to the same locked door.
Follow the two findings to their joint conclusion and the report’s politest sentence becomes its most damning: the system’s failures are not neutral. A waiting list is indifferent to you; a system that holds a legal exit and doesn’t mention it is not indifferent. Every person not told about Right to Choose was a person whose statutory entitlement was cheaper unexercised. The Ombudsman stops short of saying the quiet part — that the information gap runs in the direction of the system’s financial interest — but the direction of every documented failure runs the same way, and patients noticed before the regulator did. That is why the complaints tripled.
The structural chapter : a regulator's gap, a missing tariff, and standards that do not exist
The report’s challenges chapter is the part almost nobody will read, and it contains the findings with the longest half-life. Three stand out.
First, a regulatory gap that should astonish anyone hearing it for the first time: providers offering only diagnostic assessment — no treatment — are not required to register with the Care Quality Commission at all. The assessment market that Right to Choose and private demand have built has a segment operating outside the regulator’s perimeter, entirely lawfully, because the registration rules were written for a world where diagnosis and treatment lived in the same building. Whatever view one takes of independent providers — and this corpus has defended their role in a failing system — a market segment with no regulator is a market segment where quality is a rumour.
Second, there is no national accreditation standard for what an ADHD or autism assessment must contain. NICE’s guideline covers the clinical criteria but is silent on service standards; the NHS ADHD Taskforce recommended competency frameworks; nothing yet binds. The practical meaning: “an assessment” is not one thing, and the argument about whether independent-sector assessments are rigorous is unresolvable in principle, because there is no standard against which any assessment — NHS or independent — can be measured.
Third, the data doesn’t exist. There are no national waiting-time standards for ADHD and autism pathways of the kind that exist for elective surgery or cancer, and the report describes data collection as fragmented to the point where national demand cannot be accurately quantified. The NHS Confederation, responding the same week, called the cost trajectory unsustainable and pointed at the absence of a national tariff — meaning boards don’t even have a standard price for the assessment they’re commissioning. A system with no waiting standards, no service standards, no tariff and a partly unregulated provider market is not a system experiencing pressure. It is a market that was never designed, being audited for the first time.
What the Ombudsman can ask for, and what a complaints file cannot see
The report’s asks are proportionate to an Ombudsman’s remit: national guidance from government, clarification of Right to Choose, support for integrated care boards to commission coherently. All correct, all necessary, none sufficient — and the report lands in a week when the bigger vehicle is already moving, with the independent review of ADHD and autism prevalence and services chaired by Peter Fonagy (see his interim report here) expected imminently. If that review reports this autumn, this document becomes its evidential floor: the complaints record that makes “the system is working adequately” an unavailable conclusion.
But the deepest limit of the report is methodological, and it deserves stating because nobody else will: a complaints analysis can only see the people with the capacity to complain. Navigating a PHSO complaint requires literacy, executive function, persistence across months, and the residual belief that institutions respond to process — the exact resources that ADHD and autism can tax, and that the most failed people have exhausted. The mortality review’s highest-risk group — late-diagnosed, comorbid, unsupported — is precisely the group least represented in 3,000 complaint files, because the person nine years into an untreated spiral does not file structured correspondence with an ombudsman. The report is an audit of the system’s failures as witnessed by its most functional casualties. The true picture is worse than the worst document the state has yet produced about it — and the worst document the state has yet produced is damning.
The Ombudsman has done the part an ombudsman can do: made the failure official, dated, and citable. What happens next belongs to the Fonagy review, to government, and to whether “postcode lottery” hardens from description into anything with enforcement behind it. The corpus position stands: every structural delay documented here is, on the evidence of the mortality data, not administrative friction. It is exposure.
Citations
Parliamentary and Health Service Ombudsman (2026) — Improving access: ADHD and autism assessment and treatment commissioning — PHSO
Rahali, Y., Dennison, C. A., Thapar, A. & Thapar, A. K. (2026) — Attention deficit hyperactivity disorder (ADHD) and premature mortality: A narrative review of the literature — European Child & Adolescent Psychiatry
National Institute for Health and Care Excellence (2018) — Attention deficit hyperactivity disorder: diagnosis and management (NG87)
NHS Confederation (2026) — response to the PHSO review on ADHD and autism commissioning — nhsconfed.org
