Article 4 of 6

Max Pemberton paid £1,200 for a private assessment, received an ADHD diagnosis and rejects it. That may expose an important quality problem. One disputed case cannot tell us how accurate private assessment is as a whole — still less whether ADHD exists.

Pemberton’s personal experiment appears, at first sight, devastating.

For The Great ADHD Myth?, Pemberton underwent an online assessment and formal report costing £1,200 from a provider he chose because it appeared reputable and used an experienced consultant psychiatrist. He completed forms, a family member answered questions about his childhood, and he underwent a 90-minute interview. The outcome was combined-type ADHD and a prescription for lisdexamfetamine.¹

He does not accept the diagnosis.

If a successful psychiatrist who believes he does not have ADHD can pay for assessment and emerge with a label and controlled drug, why should anyone trust the process?

It is a fair question. It is also more limited than it sounds.

This may have been a poor assessment. It may have been a borderline judgement about impairment. Pemberton may meet formal criteria while finding the diagnosis an unhelpful way to understand himself. Without the full clinical material and an independent reassessment, readers cannot decide between those possibilities.

What the episode cannot establish is how often private diagnosis is wrong, whether a diagnosis can generally be bought, or whether ADHD is a valid clinical construct.

The strongest case against the private market

The concern deserves to be steelmanned.

ADHD has no confirmatory laboratory test. Its symptoms overlap with common experience and with anxiety, depression, trauma, sleep disorders, substance use, autism, learning difficulties and physical illness. Adult assessment must reconstruct childhood onset, sometimes decades later, when school records are lost and relatives are unavailable or uncertain.

Patients usually request an assessment for a named condition and may arrive after months of research. They may understandably hope for an explanation, treatment or adjustments. The provider is paid for the assessment and may also receive income from titration, prescriptions and follow-up. A negative conclusion can disappoint the customer while leaving their difficulties unresolved.

Those incentives do not prove dishonesty. They create conditions that require explicit standards, transparent reasoning and independent audit.

The NHS is not automatically the virtuous comparator. Years-long waits select for people able to persist, and stretched services can make both false-positive and false-negative errors. But NHS scarcity has fuelled a market in which providers, assessor backgrounds and follow-up arrangements vary. The NHS England ADHD Taskforce has called for regulation, minimum assessment standards and better data across all providers.²

The right question is not private or public? It is did this assessment meet a defensible standard, and is the provider accountable for what follows?

A checklist is not a diagnosis

Pemberton’s broader critique repeatedly draws attention to ordinary-sounding criteria. In The Spectator, after listing behaviours such as talking excessively and struggling to engage quietly, he wrote: “If everyone can have a diagnosis, then no one does.”³

The conclusion is sound only if having a diagnosis means recognising one item. It does not.

Almost everybody sometimes loses things, procrastinates, interrupts or becomes restless. ADHD is a pattern defined by persistence, developmental onset, presence across settings, impairment and exclusion of better explanations. The category’s boundary is contestable, as Article 3 explained, but it is not simply a tally of familiar quirks.

NICE guidance requires an appropriately trained specialist to undertake a full clinical and psychosocial assessment, developmental and psychiatric history, observer reports and mental-state assessment. Symptoms must cause at least moderate impairment and be pervasive in two or more important settings. Rating scales may contribute but should not make the diagnosis alone.⁴

The 2024 Adult ADHD Assessment Quality Assurance Standard recommends a systematic, semi-structured interview; explicit evaluation of impairment and competing explanations; collateral information where possible; and at least two hours for the diagnostic assessment and initial post-assessment discussion.⁵ It is a professional consensus standard rather than a legal rule, but it gives patients and commissioners something concrete against which to examine practice.

An online assessment is not inherently invalid. Video can support a detailed history, collateral evidence and careful reasoning. Conversely, an in-person appointment can be superficial. The issue is not the screen; it is the method.

What do we know about Pemberton’s assessment?

His published account supplies more detail than a headline suggests.

He describes persistent compensatory systems: multiple alarms, deliberately altered timings, a daily index card and strict control of his inbox. He reports frequently losing his phone without a particular strategy, reading the same paragraph repeatedly, interrupting, working to loud music and experiencing “chaos” without his systems.⁶ He sees these as manageable traits that have contributed to a successful and energetic life, not disability.

That interpretation may be entirely reasonable. A diagnosis should not rest on traits without significant impairment. If the assessor inferred impairment without demonstrating it, the report’s central judgement is vulnerable.

Yet his certainty is not an independent reference standard. Doctors can misjudge their own health in either direction. Achievement does not exclude ADHD: ability, occupational fit, support, anxiety, overwork and elaborate compensation can conceal functional cost. This is particularly relevant to adults who were quiet or predominantly inattentive as children, girls and women overlooked by disruptive-behaviour stereotypes, and people whose difficulties became visible only when structure fell away.

That does not mean every hard-working or highly structured person has hidden ADHD. Compensation is evidence to explore, not a diagnostic trump card.

Three interpretations therefore remain open:

1. False positive: relevant traits were present but childhood onset, pervasiveness or impairment was not, and the assessor diagnosed too readily.
2. Borderline disagreement: the evidence was real but reasonable clinicians could differ on whether functional cost crossed the threshold.
3. Criteria met, identity rejected: the assessment was defensible, but Pemberton does not find the diagnosis or disability language useful.

The film and report may narrow those possibilities. The public accounts do not eliminate them.

What did the “experiment” actually test?

Methodologically, this was an uncontrolled case involving an unusually informed participant.

Pemberton knew the criteria and the documentary’s purpose. Even without deliberate deception, people select examples and emphasise difficulties differently depending on the question they think is being asked. The assessor knew that the patient had requested an ADHD assessment, as in ordinary practice, but there was no comparison group and no independent reference diagnosis.

If he answered in a wholly ordinary way and the clinician supplied impairment not present in the history, the case reveals a serious specificity problem in that assessment. If he knowingly performed a symptom profile, it tests something else: whether a history-based assessment can detect skilled fabrication. If the report documented substantial functional cost that he subsequently reframed, the disagreement concerns judgement and meaning.

One result cannot yield a false-positive rate. It is closer to a case report than a validation study.

To estimate accuracy, researchers would need representative patients from multiple NHS and private services, including people eventually given no diagnosis or a different diagnosis. Independent clinicians, blinded to provider and payment route, would review standardised histories and collateral evidence. The study would measure agreement, alternative diagnoses, subsequent outcomes and variation between services. Even then, in the absence of a biomarker, disagreement would not automatically reveal which clinician was correct.

Pemberton’s case identifies the need for that work. It does not perform it.

“Diagnosis creep” is a legitimate concern

Diagnostic manuals change who is included. DSM-5 moved the age by which several symptoms must have appeared from before seven to before twelve. In one US population study of 12- to 15-year-olds, applying the later threshold increased estimated prevalence from 7.38 to 10.84 per cent.⁷

That numerical increase is real. Boundaries are human decisions and should be examined for benefit and harm.

The study also found that the group newly included by the later onset criterion did not differ significantly from the earlier-onset group in symptom severity or patterns of psychiatric co-occurrence.⁸ The change may have captured clinically important difficulties that were recalled or recognised later rather than simply redefining ordinary children as ill.

A broader threshold can create false positives; an unjustifiably narrow one creates false negatives. “The criteria changed” begins an evaluation. It does not complete it.

Diagnosis and disability are not synonyms

Pemberton objected when the assessing clinician described him as disabled, particularly in comparison with a colleague who had a progressive neurological condition.⁹ His discomfort is understandable. The comparison nevertheless confuses category with degree.

Disabled people do not share one severity, prognosis or need. Nor does an ADHD diagnosis automatically settle the legal question. Under section 6 of the Equality Act 2010, disability ordinarily requires a physical or mental impairment with a substantial and long-term adverse effect on normal day-to-day activities. The individual effect matters.¹⁰

Pemberton is entitled not to identify as disabled and not to use a diagnosis he finds unhelpful. That preference does not determine whether another person’s ADHD substantially affects education, work, relationships, finances, self-care or safety.

It also highlights a communication failure. Clinicians should explain what they mean by impairment and disability, rather than present either as an identity that automatically follows from a checklist.

What the case justifies

It would be a mistake to dismiss Pemberton’s assessment because his larger conclusion is unproved. A paid, high-stakes diagnosis followed by controlled-drug prescribing must withstand scrutiny. His account supports a practical agenda:

• enforceable minimum standards across NHS and independent providers;
• transparent assessor training, supervision and accountability;
• explicit developmental history, cross-setting impairment and differential diagnosis;
• collateral information where reasonably available, with its absence documented rather than automatically disqualifying;
• a reasoned report that distinguishes evidence, uncertainty and judgement;
• safe baseline assessment, titration, monitoring and continuing care; and
• routine publication of diagnostic yield, audit results and outcomes.

Support for ordinary functional difficulties should also be less dependent on acquiring a medical label. When extra time, coaching or workplace flexibility is available only after diagnosis, systems add pressure to the threshold and make the assessment carry more than a clinical decision.

The warranted conclusion is neither “the clinic must be right” nor “private ADHD is for sale”. It is that present systems do not provide enough transparent quality assurance to know how often assessments go wrong or in which direction.

Pemberton may have exposed a poor assessment. He has certainly exposed a quality-control question.

That question deserves representative data, not insinuation about every privately diagnosed patient — and not a title-sized conclusion about whether ADHD is real.

---

Series disclosure: I am a doctor diagnosed with ADHD in adulthood and take lisdexamfetamine. This is context, not evidence; the clinical claims in this series should stand or fall on the sources cited.

Sources and further reading

Series navigation

← Previous: Article 3 — Neurodevelopment, biology and social context · Series index · Next: Article 5 — Medication: benefit, harm and choice →