专家评议Fonagy独立报告：ADHD与自闭症应否以诊断作为支持门槛

一份报告，两种焦虑

由Peter Fonagy主持的独立报告《心理健康状况、ADHD与自闭症》于英国时间10月9日发布，Science Media Centre随后汇总了多位研究者的书面反应。这些文字不是新实验，也不是对报告的同行评审，而是专家在初读后给出的判断与保留意见。它们共同围绕一个问题：当评估与支持资源长期不足时，是否还应把正式诊断当作获得帮助的入口。

报告的核心主张是把系统从“依赖诊断”转向“需求主导加诊断”。支持不应以拿到某个诊断标签为前提，诊断仍应保留，但只在必要时跟进。多位专家认可这一方向，认为它承认了贫困、住房不稳、歧视与不可及的环境也会制造痛苦，并承认自闭症与ADHD不是需要被预防的状况。

诊断是不是可以被替代

最集中的分歧在于：不依赖诊断的支持，是否会在实践中变成不提供诊断。杜伦大学Monique Botha指出，报告没有给出有力证据证明心理教育足以替代诊断。心理教育、同伴支持和实用建议可能有价值，但“有帮助”不等于“能提供与及时、准确诊断相同的益处”。她担心报告虽写明等待期间的支持应是通往诊断的桥梁，但建议把快速专科评估优先给功能损害严重或出现恶化迹象的人，而需求看似较轻者则在期限未定的等待中获得支持，桥梁可能变成永久安置点。

Botha还提出一个评估上的难点：功能损害不完全是个人身上的特质，它产生于个人、环境要求与可用资源之间。一个人可能因为有家庭支持、自费购买帮助、掩饰困难或牺牲生活其他部分而维持工作或学业，外在参与程度很难说明这种状态是否可持续。若按可见损害排序，可能重现报告想改变的局面——人要到接近危机时才被看见。

谁被漏掉，谁被过度诊断

伦敦国王学院Gavin Stewart认为，自闭症与ADHD的潜在患病率没有明显改变，改变的是识别能力；现有最佳人群证据显示两者仍被大幅低估，尤其是老年人、女性与女孩，以及少数群体社区。他欢迎报告区分诊断不足、过度诊断与误诊可以同时存在。

Andrea Danese进一步指出，报告拒绝把过度诊断与诊断不足做成二选一：在ADHD与自闭症上，部分年龄组的记录诊断分别增长六倍与二十倍，已接近或超过人群估计；而抑郁与焦虑方向相反，多数年轻人从未到达专科服务。他引用报告中的STADIA试验称，转介到社区心理服务的儿童中约八分之一在一年内获得记录诊断，尽管三分之二接受评估者筛查显示常见精神障碍可能性高。他还强调，报告承认自2017年以来没有全国调查用完整诊断方法评估儿童，因此争论核心问题目前无法用现有数据定论。

Rachel Moseley则担心报告仍把记录诊断上升与潜在患病率稳定作对比，容易给人“需求可能是虚假的”印象。她认为自闭症与ADHD历史上长期被低估，当前需求上升很可能主要是补课效应，而人群患病率估计本身很可能偏低。

对数字与AI工具的谨慎

Anita Thapar在评议ADHD部分时指出，报告建议与她2025年主持的独立ADHD工作组建议基本相同，参与者大多相同，这让她质疑为何要重复这项工作，并提到重复宣布报告给有ADHD亲身经历者和等待名单上的人带来痛苦，也放大了媒体猜测。

她对数字与AI建议的清晰度表示担忧：报告后来限定等待名单需有临床监督复核，数字产品必须经过验证并达到与其他干预相同的标准。Danese补充，报告对数字与AI持适当谨慎，指出有指导的数字干预优于无指导干预，后者难以维持参与；报告把自身数字建议评为中等或新兴证据，而非已确立，并明确AI不应成为对风险人群的唯一回应。他举出一个被评估的分诊系统：技术本身可靠，但阈值设得过于围绕危机，结果大多数有明显损害的人落入同一个未区分等待池。工具表现取决于它嵌入的服务规则。

分歧背后的共同点

Uta Frith认为报告咨询过广，结果是在关键问题上骑墙，多数建议是愿景而非可操作方案，并失望于报告没有强调改进诊断工具的科学基础。Peter Tyrer则质疑报告对近年诊断差异的解释不足，认为自闭症谱系范围扩大本身就会带来更多诊断，并批评报告未讨论人格病理。他提出部分所谓障碍可能被年轻人用来回避工作的说法，这一说法在材料中没有提供实证支持，属于其个人判断。

若从这些评议中找共同点，多数专家都同意：支持应更早到达，不应等到危机；诊断与支持不应互相替代；扩大非诊断支持路径的同时，必须保留及时评估的可及性，并用实际结果检验新安排是否让人更早获得合适帮助。分歧在于，在资源有限的系统里，谁先被评估、谁承担等待的代价，以及现有证据是否足以支撑改变评估门槛。

Scientists comment on the independent review into mental health conditions, ADHD and autism: final report and recommendations.

Prof Anita Thapar, FRCPsych, PhD, FMedSci, CBE, Cardiff University, said:

“This is a very long and detailed report, and I commend Peter Fonagy and colleagues for how scholarly it is. It will take a few days of reading to ensure I have identified the main points and not missed anything crucial. I will focus on ADHD here.

“Essentially, in relation to ADHD, the recommendations are the same as those put forward by the Independent ADHD Taskforce in 2025 that I chaired. This is unsurprising given that most of the scientists involved are the same. That is good but at the same time does raise the question as to why the work on ADHD needed to be repeated. Apart from delaying changes, the announcement of another report created distress to many with lived experience of ADHD and those on waitlists and hugely increased media speculation.

“These same recommendations across both reports include: adopting a needs-based approach but recognising that will not substitute the need for diagnosis. Both reports highlighted the need to ensure support in different sectors was not contingent on receiving a diagnosis (highlighted by both the ADHD Taskforce and dfE Neurodivergence Task and Finish group in 2025).

“Both reports highlighted the importance of ensuring that support crosses mental health conditions and neurodivergence rather than diagnosis alone or a single diagnosis.

“Also both emphasised early support and intervention, early years and school support, cross sector, whole system changes from government to locality, regulation of ADHD clinical assessment and treatment providers by CQC among others.

“One concern on initial reading is that the recommendations on digital and AI approaches are not entirely clear and could be misinterpreted as specific examples are not provided. Later, the report does qualify that wait lists need review with clinical oversight and that any digital products used must be validated and shown to meet the same standards as other interventions. This is important given the potential of missing those who need urgent support and managing risks such as suicide in those who may not be first recognised as risky and the lack of robust evidence on how best to safely triage. Another concern is that it is not completely clear on first reading that increased NHS capacity is a priority alongside systems change and support in other sectors.”

Dr Gavin Stewart, British Academy Postdoctoral Research Fellow – Re:Spect Lab, Institute of Psychiatry, Psychology and Neuroscience, King’s College London, said:

“The Fonagy report is timely and sets out clear recommendations for the way forward. It rightly recognises that our understanding of autism and ADHD has grown considerably in recent years, meaning more people are now seeking assessment. There is little evidence that the underlying prevalence of autism or ADHD has changed; what has changed is our ability to recognise people who may previously have been overlooked.

“Underdiagnosis, overdiagnosis and misdiagnosis can all occur at the same time, and the report makes an important distinction between them. Our best population level evidence strongly indicates that autism and ADHD remain substantially underdiagnosed, particularly among older adults, women and girls, and people from minoritised communities. I strongly welcome the report’s recognition of these inequalities, and hope these groups become a focus in future.

“Too many people still face long waits and barriers to assessment and support, which can allow difficulties to escalate into crisis. I welcome the move towards support based on need, rather than making a diagnosis the gateway to help, though this will take sustained investment, workforce growth and careful implementation. People should be able to get the support they need when they need it, whether or not they have a diagnosis.”

Dr Monique Botha, Associate Professor in Social and Developmental Psychology, Durham University, said:

“There is much in this report that I support. People should receive help before they reach crisis, and access to practical support, adjustments and appropriate mental healthcare should not depend on first securing an autism or ADHD diagnosis. I welcome its recognition of poverty, insecure housing, discrimination and inaccessible environments as contributors to distress. I also welcome its acknowledgement that autism and ADHD are not conditions to be prevented, and that timely diagnosis and meaningful support should proceed alongside one another.

“My greatest concern is that there is no robust evidence presented that psychoeducation is an adequate replacement for diagnosis. Psychoeducation, peer support and practical advice can be valuable, but evidence that they help people is not evidence that they provide the same benefits as a timely, accurate diagnosis. Before changing access to assessment, we need to understand the consequences across people’s lives, including their self-understanding, access to appropriate treatment, material security and the care they receive as they age.

“The report explicitly says that support while waiting should be a bridge to diagnosis rather than a substitute for it. That commitment matters. However, its recommendations give rapid specialist assessment priority to people with severe functional impairment or signs of escalation, while offering those with apparently milder needs support during an unspecified wait. Without protected access to timely assessment, adequate funding and accountability, there is a substantial risk that the bridge becomes a permanent holding place.

“Functional impairment is not simply a trait located within an individual. It emerges through the relationship between a person, the demands placed on them, and the resources and support available to them. Someone may remain in work or education because they have substantial family support, pay privately for help, mask their difficulties, or sacrifice almost every other part of their life. Outward participation tells us very little about whether that situation is sustainable. Nor should people have to lose the support that keeps them functioning before their need for assessment becomes sufficiently visible.

“The report recognises parts of this problem, including the personal cost of maintaining participation. My concern is whether that understanding will carry through into decisions about who receives an assessment. In an underfunded system, prioritisation by visible impairment can reproduce precisely the arrangement the report wants to change: people receive recognition once their circumstances have deteriorated enough to demand an urgent response. This replicates the “issues that we currently face which is that people only get help when deep in crisis. Diagnostic waiting lists are already full of people who are experiencing rock bottom. 

“Diagnosis and self-understanding can themselves contribute to preventing that deterioration, and we need better preventative care that recognises that neurodivergent people deserve fundamentally, a chance to thrive. Understanding why particular environments are overwhelming, why certain demands repeatedly lead to exhaustion, or why previous treatment has not helped can enable people to make informed decisions and seek appropriate adjustments. Ambiguous psychoeducation might end up being more confusing, and less accessible because we aren’t giving access to evidenced based interventions for the kind of diagnoses these people might meet the criteria for. The report acknowledges these benefits from diagnosis, but they need to carry meaningful weight in assessment policy, including for people who are not currently in crisis.

“This is also a question about ageing. A person’s needs do not end when they leave school, complete a course of psychoeducation or appear to be managing in employment. Their circumstances change. Family support may disappear, health may deteriorate, and they may eventually need support at home, hospital care or residential care. We need to ask whether the understanding offered today will remain usable by the services responsible for their care decades later. I am particularly concerned that delaying diagnosis could make later diagnosis harder to obtain. If a child, who clearly is autistic or has ADHD but is being supported by knowledgeable parents (often neurodivergent themselves) who constantly scaffold, and re denied recognition because they are managing to live well, it also then skews perceptions of the kinds of lives available for neurodivergent people, ignores the invisibilized burden on neurodivergent parents who are working tirelessly, and delays access to the kind of diagnosis their child will need as they age out of the care of their parents. 

“As people accumulate mental and physical health diagnoses, there is a risk that their longstanding neurodevelopmental needs become overshadowed by other explanations for their difficulties. Anxiety, depression, trauma or age-related changes may become the focus, while the underlying pattern remains unrecognised. Developmental histories and people who can help establish them may also become less accessible over time. Deferring assessment now cannot be assumed to preserve an equivalent opportunity for assessment later.

“A diagnosis does not automatically guarantee good care. But without timely recognition and a durable account of someone’s needs, I worry that those needs will not be translated into adequate practice. Sensory, communication and predictability needs must be understood and acted upon when someone requires care, including when they are less able to explain or advocate for themselves. A short-term support offer needs to be evaluated against that lifelong responsibility.

“The report also acknowledges that it cannot establish overdiagnosis on scale. That uncertainty should constrain the conclusions drawn from increasing diagnosis rates. More recorded diagnoses do not, by themselves, demonstrate inappropriate diagnosis, just as fewer diagnoses would not demonstrate better wellbeing. Shorter waiting lists could reflect improved access, but they could also reflect people being redirected without their needs being adequately understood. The concern that the risk of over diagnosis is increasing is not fully evidenced and doesn’t logically follow and I worry it will whip up more hostile rhetoric.

“I support expanding the routes through which people can receive help. That expansion must come with greater choice and security, including continued access to diagnostic assessment. Agency and self-determination are so important. Any change should be judged by whether people receive appropriate support sooner, retain meaningful control over their care, and experience better outcomes throughout their lives. The burden of proof lies with demonstrating that the proposed alternative protects those outcomes. People should not bear the long-term consequences of an untested assumption that psychoeducation can adequately replace diagnosis.”

June Brown, Professor of Clinical Psychology, Institute of Psychiatry, Psychology and Neuroscience, King’s College London, said:

“I welcome this very thorough, bold and ambitious report.

“I welcome the distinction made between Diagnosis v need. This is a crucial issue in the disagreements about who should and should not get support for ADHD/autism etc. Related to this, I also the distinction made between needs led support and diagnosis led treatment; the way in which decisions are going to be made is very strongly focussed by the WHO ICF system. But it does change how we see people with mental health issues in a less ‘medical’ way. However, this is going to take time for staff to learn, and an interim phase may need to be phased in.

“I have some comments about other specific recommendations to do with funding and staffing. I like the recommendations for self-referral youth hubs so that young people can refer themselves. However, there is an issue about the staffing needed for this because the Australian government has put a lot of money into this. This is part of their Early Intervention system which I applaud. Neighbourhood multidisciplinary teams I would like to see more details about the co-ordination of the services in the community; neighbourhood multidisciplinary teams could be very good, but they will need good skilled staff and co-ordination. Mental health support teams I agree that mental health support teams who work in schools are absolutely invaluable to this endeavour of reaching people earlier rather than waiting for problems to develop before reaching out for help. The need to provide evidence-based interventions is very important.

“Finally, the report recognises there needs to be better education of the general public about mental health problems – both needs and diagnosis. There is mention of digital and AI information, but I think that in in-person forums for public mental health education are needed as it is easy to misunderstand what are problems and what are not.”

A spokesperson for NICE, said:

“We welcome the report’s focus on improving access to NICE-recommended care for children, young people and adults. We are now considering what further updates to our guidance relating to autism and ADHD may be needed in light of the findings.”

Dr Hannah Bowers, Lecturer at the Primary Care Research Centre, University of Southampton, said:

“Support without a diagnosis is a welcome idea, and many people will benefit from getting help sooner. In our interviews and public involvement work with autistic and ADHD adults, people said a diagnosis helped them understand themselves and explain their needs to others. If we prioritise diagnostic assessments for people with higher needs, those with mild to moderate needs might end up with support in place of a diagnosis instead of alongside. A shift towards GP surgeries managing ADHD and autism raises further questions about how primary care can meet these needs and offer support. Our participants described rejected referrals, little support, and therapies that did not fit their needs.”

Professor Subodh Dave, President of the Royal College of Psychiatrists said: 

“Millions of people have had their lives turned upside down by poor mental health, including illnesses and neurodevelopmental conditions that are often severe, disabling and overlapping. The report makes it clear that we must reduce unacceptably long waiting lists for high-quality assessments and improve access to specialist treatments that are too frequently unavailable if we are to address this.  

“More children and young people are suffering significant distress because of a constantly changing world they have almost no control over. Those showing initial signs of distress shouldn’t need a diagnosis to access basic levels of support at school, work or within their communities.

“We hope Professor Fonagy’s review will drive renewed focus on mental health at every level of society in England, particularly for people with severe and enduring mental illnesses. We need mental health services that are responsive, connected and consistent if we are to support people to be healthy and active, which must be backed by an expansion of the workforce and investment. Those with the most complex needs must have timely access to psychiatric expertise.”

Dame Uta Frith, Emeritus Professor of Cognitive Development, UCL, said:

“The report had a difficult task. It consulted too widely and hence the result is a determined sitting on the fence. The report skirts the question of why the huge rise in demand for ADHD and ASD diagnosis is not found for other psychiatric conditions. I was disappointed not to see any emphasis on improving the scientific basis for the conditions in question, in particular the validity and reliability of current diagnostic instruments.

“Most of the recommendations are aspirational rather than practical. The radical system change that is recommended seems utopian as it does not consider the limitations of political and societal support. The emphasis on needs is important but the call for a needs- plus diagnosis- based system does not solve the problem of how to assess needs and prioritise them.

“I applaud the call for a ban on advertising for private assessments and instituting strict checks of the qualification of assessors. However, even the currently approved diagnostic procedures for ADHD and ASD, with their reliance on purely subjective reports, need to be critically re-evaluated.”

Dr Alex Pike, Lecturer in Mental Health Department of Psychology, University of York, said:

“The Fonagy review into mental health, ADHD and autism is nuanced and driven by a range of evidence, which is carefully considered and categorised to inform the language used in presenting the evidence. Some of the key points made in the review sound purely semantic, but they have significant implications for topical debates at the moment: namely, the authors distinguish between diagnosis, function, distress, and need. The most important advance in this report is to suggest that diagnosis should no longer be the ‘key’ needed to unlock support, but instead that individual needs should be assessed in a holistic manner, with diagnosis following only if necessary. They anticipate this will reduce waiting lists: however, this may just shift the name of the assessment people are waiting for. An important point is the recognition that the social landscape has changed – increases in poverty, financial insecurity, and political instability, and greater evaluation and comparison, may explain rising distress. The review also notes that privately provided ADHD/autism assessments should be regulated by the CQC and should not be able to advertise directly to the public, which I think will be important to maintain trust in the systems providing them.”

Prof Peter Tyrer, Emeritus Professor of Community Psychiatry, Imperial College London, said:

Do you broadly welcome the report’s findings and recommendations? if so why, and if not, why not?

“The findings are generally balanced but explanation of the differences between systems not always comprehensive and believable.”

Do you think the report gets the balance broadly right on the topical issue of whether we have over-diagnosis, under-diagnosis or neither?

“As it covers all areas and gives different explanations it is largely balanced (but see below).”

Is this a good quality report based on solid scientific evidence? Are the conclusions backed up by solid data? 

“There is insufficient information about the differences in diagnosis that have occurred in recent years. In particular, autism now covers a much wider spectrum than previously and so it is hardly surprising that it is diagnosed more commonly.  The same applies to ADHD to some extent. A key part of the report states that ‘current arrangements do not always provide the most effective or efficient way of identifying and responding to need. In particular, where access to support depends heavily on formal diagnosis, demand for diagnostic assessment can increase more rapidly than systems are able to respond, generating long waiting times, duplication of assessment and pressure on specialist services.’  Attribution of a diagnosis offers the practitioner an easy way of disposing of a problem that might be best deal dealt with by other means. such as occupational guidance, family intervention and training. 

“I am especially interested in the prevalence of personality pathology and there is no mention of this anywhere in the report. The existing evidence is that personality problems also exist on a spectrum and very few people can be described as having a normal personality^[1]. Virtually all the people with ADHD and autism also have personality problems and most of these are best dealt with by non-medical approaches involving adapting the environment to the person’s needs rather than intervening medically.  By giving a diagnosis such as autism or ADHD an unnecessary degree of pathology is given to problems that in the past would be regarded as part of the normal spectrum of existence.  Current thinking over-pathologises normal variation and is particularly prevalent in primary care.

How does this work fit with the existing evidence? Does it take us further than previous reports/assessments of the rising rates of mental health distress ADHD and autism in useful and important ways? 

“The report shows that there is a mismatch between the epidemiology of these conditions and their frequency in practice.  One of the suggested reasons for the increase in prevalence of all these disorders is the wide use of social media and the sharing of mental health knowledge that was never present in the past.  It is a sensitive subject, but the authors of this report have completely ignored the possibility that at least some of these alleged disorders are created by young people as a deliberate method of avoiding work. In my discussions with people involved in the assessment of NEET people there is a widespread belief that the attribution of a diagnosis of autism or ADHD is perceived as a much better outcome than working, not least with ADHD, as treatment involves taking psychostimulants that even in normal people are perceived as pleasurable.   Peter Fonagy, in a recent television interview, was quite unequivocal in stating that this increase in pathology was real, but there is nothing in the report that can confirm this view.  Most of the official bodies have bent over backwards in emphasising the genuineness of these complaints and I think this is because they fear a backlash if they suggest that some of the pathology is fictitious.

“I have been involved in the World Health Organisation’s revision of the ICD-11 diagnostic system (not yet implemented in the UK) and was the Chair of the working group involved in the revision of personality disorders, which removed all categories of the previous diagnostic system and placed all personality pathology on a single dimension. The DSM system is now following suit. I personally believe, but you might expect me to say this in view of my orientation, is that a large proportion of these problems are best dealt with by adjusting the personality to situational needs (Tyrer P (2025) Adaptive Solutions to Personality Disorders: Treating Patients with Nidotherapy, Routledge) and am currently helping the training of those involved in the management of NEET people by asking them to take account of personality functioning in terms of both strengths and weaknesses. These are early days but those involved feel that this input has been very helpful.”

[1] Yang et al., 2010, British Journal of Psychiatry. 197(3):193-199

Prof Miranda Wolpert, Director of Mental Health at Wellcome, said:

“The review is an important contribution to understanding how we can better respond to the growing burden of mental ill health. Real progress will come from scaling evidence-based approaches for those who need it most. This will mean people receive help earlier and reduce long-term impacts on their education, employment and health. We have an opportunity to build and implement solutions that will change lives.”

Dr Rachel Moseley, Principal Academic in Psychology, Bournemouth University, said:

“The report advocates for a “fundamental shift from a diagnosis-dependent system to a needs-led plus diagnosis system”. Researchers like myself, whose work highlights severe mental illness and suicidality in undiagnosed populations, welcome recognition that our educational, health and social care sectors must have the flexibility and skills to rapidly offer autism- and ADHD-related support to people without diagnoses – lives quite literally depend on it. However, within the subtext of the report is the argument that specialist capacity must be “protected” for “those who need it most”; those with “the most severe, enduring and complex needs”. It is inarguable that neurodivergent people with obvious, complex difficulties need and deserve support and diagnosis. However, we have observed a growing trend calling into doubt the legitimacy of needs in autistic and ADHD people whose differences are less extreme – people who attend mainstream schools, sustain employment for prolonged periods, yet suffer excess morbidity and mortality, particularly through suicide. There is, as such, a threat that diagnosis and support will be rationed to the detriment of these groups, who find a diagnosis “life-saving” for the explanation it gives them, but who also have significant need for support. 

“While I agree that continued focus on simplistically posed questions about over- and under-diagnosis is unhelpful and potentially harmful, the report still fails to recognise just how inaccurate our prevalence estimates for autism and ADHD are likely to be. It continues to set a contrast between increases in recorded diagnoses and demands for assessment on one hand, and stable underlying population prevalence on the other – and by doing so, gives the impression that rising diagnostic incidence and demands for assessment may be spurious, without sufficiently recognising the fact that our estimates of underlying population prevalence are almost certainly underestimated. Both autism and ADHD have been historically under-recognised, and as such, rising demand for assessment and greater numbers of recorded diagnoses are likely to largely reflect a catch-up effect – rates would indeed, as the report suggests, be expected to stabilise if we were able to ‘catch up’ on all those people who’ve been missed, but we are far, far from this point, and the report still fails to reflect this.”

Prof Andrea Danese, Professor of Child & Adolescent Psychiatry, Institute of Psychiatry, Psychology & Neuroscience, King’s College London, and Consultant Child & Adolescent Psychiatrist, National & Specialist CAMHS, South London and Maudsley NHS Foundation Trust, said:

“I welcome this report. It takes a polarised public argument and replaces it with a careful reading of what the data can and cannot show. Its central distinctions — that population prevalence is not administrative prevalence, that distress is not disorder, that diagnosis is not need, and that support is broader than treatment — are the right ones, and they have been missing from the debate.

“On the balance between over- and under-diagnosis, I think the report gets it broadly right, and importantly it refuses the binary. Both can be true in the same system, in different places. The concern about diagnoses outpacing need rests mainly on ADHD and autism, where recorded diagnoses have risen more than six-fold and twenty-fold respectively in some age groups and now approach or exceed population estimates. For depression and anxiety the position is reversed. Most young people with these conditions never reach specialist services — the National Audit Office estimates only around a quarter of children with a probable mental health condition are accessing NHS-funded care — and reaching a service does not guarantee recognition. The STADIA trial, cited in the report, found that around one in eight children referred to community mental health services received a recorded diagnosis within a year, although two thirds of those assessed screened at high probability of a common mental disorder.

“The report is also right that the rise is real rather than an artefact of disclosure. The adult survey uses a structured diagnostic interview applying clinical criteria including duration and impairment, and shows common mental disorders among 16–24-year-olds rising from around one in six in 1993 to one in four in 2024. The pattern of change is selective — worry, anxiety, depressive ideas and panic have risen, while irritability has fallen and conduct problems have remained stable. Greater willingness to report would not produce that pattern.

“On quality, this is a serious piece of work. It triangulates across population surveys, administrative data and lived experience, states explicitly what each can and cannot answer, and grades its own recommendations by evidence maturity — strong, moderate or emerging. That candour is unusual and welcome. Its most important admission is a gap, not a finding: no national survey has assessed children using full diagnostic methods since 2017, so the question at the centre of the public debate cannot currently be settled with the data we collect. The recommendation to commission those surveys should be acted on first, because without them we will be having this same argument in five years.

“The report advances the field by shifting attention from whether too many diagnoses are given to whether the right people are being identified and helped in time. That is the more useful question. But the move to a needs-led plus diagnosis system will only work with a workforce large enough to assess people properly and to act in good time for those placed lower down the queue. Structured assessment and disorder-specific treatments change lives. What is missing is the capacity to deliver them at scale.

“On digital and AI-enabled care, the report is appropriately cautious. It notes that guided digital interventions outperform unguided ones, which achieve little sustained engagement, and that regulated digital therapeutics now carry conditional NICE recommendations. It grades its own digital recommendations as moderate or emerging rather than established, insists that tools making clinical claims be evaluated to clinical standards, and states that AI should never be the sole response to someone at risk. It also gives a telling example: one evaluated triage system was technically sound but of limited operational value because its thresholds were set so narrowly around crisis that most people with significant impairment ended up in the same undifferentiated waiting pool. That is the right lesson — the performance of these tools depends on the service rules they sit inside, and models trained on historical clinical data can reproduce existing inequalities.”

Thea Stein, Chief Executive Nuffield Trust Health, said:

“With compassion and rigour, this report brings much-needed focus to the growing levels of distress we see across the population, particularly among young people. It makes a powerful case that we have asked too much of diagnosis and simultaneously done too little to provide the practical support needed by people experiencing distress. 

“Moving towards a system based on people’s needs is sensible, but it can only work if support is actually there for people who need it. The knotty question for policymakers is how we decide who needs clinical care and diagnosis and who can be better helped through support in work, school and wider society.  

“For ADHD and autism, there is mounting evidence of how hard it is to distinguish symptoms, functional impairment and support needs, and the report raises concerns about the consistency and quality of diagnostic assessments for these conditions. Our own analysis published last month adds to this picture, showing how the historically clear link between deprivation and receipt of disability benefits for ADHD and autism is weakening. This raises difficult questions about how people’s circumstances and expectations shape the recognition of disability and the support they receive.  

“The report also shines a light on whether growing demand for autism and ADHD support are being met in the right way, including by private providers. People who need support are facing a wild west of private companies trying to fill the gaps, and it’s very difficult to determine which ones are offering good care.  

“Today’s report marks an important step in moving the debate on from unhelpful caricatures like the ‘snowflake generation’. But change comes with a price tag. The over 400 pages of findings and recommendations are at risk of gathering dust if the government doesn’t work quickly to make hard choices and commit the funding to invest in the new service models Fonagy has outlined.” 

‘Independent review into mental health conditions, ADHD and autism’ led by prof Peter Fonagy was published at 00:01 UK time on Friday 9th October.

Declared interests

Prof Anita Thapar: “Anita Thapar served as Chair of the NHS-England ADHD Taskforce and was co-chair of the previous Welsh Government Neurodivergence Ministerial Advisory Group.”

Dr Gavin Stewart: “Not involved in the report in any way.”

Dr Monique Botha: “none.”

Dame Uta Frith: “I have spoken and written about possible reasons for the rising diagnosis of ASD. I am known for my critical stance towards the expansion of the autism spectrum and the lowering of diagnostic thresholds.”

Dr Alex Pike: “Dr Alex Pike is a Council Member for the British Association of Psychopharmacology, and has held funding from the Academy of Medical Sciences and the Wellcome Trust.”

Prof Peter Tyrer: “I was Chair of the ICD-11 personality disorders revision group and am also the Chair of a charity supporting environmental change as management of mental illness (NIDUS-UK reg charity number 1154650).”

Prof Miranda Wolpert: “Miranda works part-time as a professor in Peter Fonagy’s department at UCL.”

Dr Rachel Moseley: “I wrote a paper on this topic (under review), examining concerns about overdiagnosis in relation to prevalence estimation and historic under-recognition. This was written as an academic working in this space, in collaboration with other experts in autism and ADHD.”

Prof Andrea Danese: “Prof Danese received funding from UNICEF (UKR/PCA2025932/PD2025718); funding from the NIHR Biomedical Research Centre: Maudsley (NIHR203318); consulting fees from Yale University’s Child Study Center and the University of Lausanne; travel costs support from ESCAP, RCPsych Faculty of Child & Adolescent Psychiatry, and Nadija Children’s Hospital and Research Institute; and he is the General Secretary of the European Society for Child & Adolescent Psychiatry (ESCAP), Chair of the Advisory Committee for the Canadian Institute for Advanced Research (CIFAR) Child & Brain Development Programme, and the Academic Secretary of the Association of Child & Adolescent Mental Health (ACAMH).”

Thea Stein: “Thea Stein was a member of the Government’s ADHD taskforce and sat on two topic groups formed as part of the Fonagy review looking at autism and ADHD prevalence and service response.”

For all other experts, no reply to our request for DOIs was received.