The introduction is also posted on Spotify as a podcast by “Gerry at The Health Equation”.

You can search Spotify for “Gerry at The Health Equation”, or use the link below:

https://podcasters.spotify.com/pod/show/gerrygaj

Please follow this show on Spotify. It really helps!

Gerry Gajadharsingh writes:

“I recently wrote about the increasing tendency towards self-diagnosis and what might be described as the liberal application of diagnostic labels such as ADHD, PoTS, hypermobility and chronic fatigue. The article below provides an interesting companion to that discussion because it concerns autism — and comes from one of the scientists who helped shape our modern understanding of the autistic spectrum.

Professor Uta Frith has spent much of her extraordinary career studying autism. What makes her present argument particularly interesting is that she is prepared, at the age of 85, to question some of the thinking that she herself helped to develop. Her concern is that the concept of an ‘autistic spectrum’ has become so broad that, in some circumstances, it risks losing diagnostic meaning.

There is a very important distinction here. Frith does not argue that people seeking an autism diagnosis are necessarily well or that their difficulties are imaginary. Her argument is more subtle: something may genuinely be making life difficult for them — but it may not necessarily be autism.

That is a distinction I find extremely important clinically. Medicine sometimes falls into a rather unhelpful binary way of thinking: either a patient has the diagnosis being considered, or there is supposedly ‘nothing wrong with them’. Clinical reality is rarely that simple. Someone can have genuine difficulties with concentration, social interaction, sensory processing, anxiety, relationships, fatigue or emotional regulation without all of those experiences necessarily representing one particular neurodevelopmental disorder.

The responsibility of the clinician is therefore not simply to validate or reject the patient’s preferred diagnosis. It is to ask: what is the best explanation for this person’s difficulties? That requires differential diagnosis and being prepared to consider several competing explanations before deciding which is most likely.

One of the recurring problems in clinical reasoning is premature diagnostic closure. This occurs when a plausible diagnosis is identified and the search for alternative explanations effectively stops. Once a label has been applied, subsequent information tends to be interpreted through it: poor concentration becomes ADHD; social difficulty becomes autism; dizziness becomes POTS; pain becomes fibromyalgia; fatigue becomes chronic fatigue syndrome. Yet each of those symptoms or experiences may have multiple potential contributors.

The diagnosis may ultimately prove entirely correct, but it should be the result of the reasoning process rather than the starting assumption.

This is one of the central themes of my forthcoming book, Clinical Reasoning for Complex Patients – The Integrated Regulatory Systems Framework (IRSF). Diagnosis remains essential, but good clinical reasoning requires us to resist premature closure: to consider competing explanations, examine the interaction between different physiological systems, and continually ask whether the diagnosis we have made still provides the best explanation for the person in front of us.

Another fascinating aspect of Professor Frith’s argument is her question about why some diagnoses have become particularly attractive. This should not be interpreted cynically. For someone who has struggled for years, receiving a diagnosis can be enormously relieving. Suddenly there is an explanation. Past difficulties begin to make sense. There may be access to treatment, support and reasonable adjustments and, perhaps most importantly, the person may stop blaming themselves.

But once a diagnosis provides a coherent explanation for someone’s life, it can become difficult to question — even when subsequent evidence suggests that the explanation may be incomplete. The diagnosis has moved beyond medicine and can become part of identity.

Autism provides another interesting example of how dramatically diagnostic patterns can change.

A large UK population study examining primary-care records reported that the incidence of recorded autism diagnoses increased by 787% between 1998 and 2018. The increase was particularly marked among adults and women. More recent Government evidence suggests that the proportion of people in England with a recorded autism diagnosis increased from around 0.1% in 1998 to 1.33% by 2023/24.

This does not mean that autism itself has suddenly become eight or ten times more common. Indeed, much of the increase almost certainly reflects something positive: greater awareness, broader diagnostic recognition, better identification of women and adults who might previously have been missed, and changes in diagnostic practice.

But an increase of this magnitude should still make us curious.

There is also a wider social context that medicine should be prepared to discuss. A formal diagnosis can now have consequences extending considerably beyond explaining symptoms. It may facilitate educational support, workplace adjustments, access to specialist services and, where functional impairment is sufficient, entitlement to disability benefits.

Disability-benefit claims themselves have risen substantially. In England and Wales, the combined number of people entitled to Personal Independence Payment or Disability Living Allowance increased particularly rapidly after 2020, rising from approximately 3.5 million in February 2020 to 5.3 million by November 2025.

These figures relate to disability benefits overall rather than autism specifically, so the two trends should not be conflated. Nevertheless, they illustrate the wider environment in which diagnostic categories increasingly have consequences extending beyond clinical description alone.

It would be wrong — and unsupported by the evidence — to conclude from this that people are seeking autism diagnoses simply to obtain benefits. Many autistic people have substantial disabilities and quite properly require financial and practical support.

But the wider context surrounding diagnosis does matter, even when nobody is consciously seeking or manipulating a particular outcome.

If obtaining a diagnostic label becomes the gateway to understanding, validation, educational provision, workplace accommodation, access to services, financial assistance and membership of a supportive community, the social significance of receiving that diagnosis inevitably increases. That does not make the diagnosis false, nor does it diminish the legitimate needs of people who are genuinely disabled. It does, however, make rigorous assessment and careful clinical reasoning even more important.

The question should therefore remain: does this individual genuinely meet the diagnostic criteria, does the diagnosis best explain their difficulties, and — perhaps most importantly — does attaching that diagnosis increase their capacity to function and flourish?

A diagnosis should open appropriate doors. We should simply be careful that it does not unintentionally close others.

Diagnostic categories are useful tools. They allow clinicians to identify patterns, communicate with one another, undertake research and select appropriate treatments. But every diagnostic category requires boundaries. If those boundaries become progressively wider, eventually people with very different underlying problems can acquire the same diagnosis. At that point the label becomes less clinically informative.

This is particularly challenging with spectrum disorders. The concept of a spectrum was enormously valuable because it recognised that biological and behavioural characteristics do not always fall neatly into binary categories. But a spectrum still needs clinically meaningful boundaries. Otherwise almost any human characteristic can eventually be incorporated within it.

We all differ in sociability. We all have preferences and routines. Many people dislike particular sounds, textures or environments. Many become intensely interested in particular subjects. Many occasionally misunderstand social situations.

The important clinical question is therefore not simply whether someone recognises individual traits associated with autism. It is whether the overall pattern, developmental history, severity and functional consequences fulfil appropriate diagnostic criteria and whether autism provides the most convincing explanation.

Social media adds another dimension that previous generations of clinicians did not have to consider. It has undoubtedly brought benefits: people with previously unrecognised conditions can discover that their experiences are shared by others and seek appropriate professional help.

But the same mechanism can encourage self-selection. Someone watches a video describing five signs of autism. Four seem familiar. They watch another. The algorithm notices their interest and supplies another twenty. Before long, what began as curiosity can become conviction.

This is not deception. It is a perfectly understandable consequence of human pattern recognition combined with confirmation bias. The difficulty arises when the subsequent clinical assessment becomes primarily an exercise in confirming the diagnosis rather than testing the hypothesis.

A good diagnostic assessment should be capable of producing either answer: yes, the evidence supports this diagnosis, or no, I don’t think this diagnosis adequately explains what is happening, and we need to look elsewhere.

If every person referred for assessment ultimately receives the diagnosis they suspected, we should at least ask whether the assessment process is sufficiently discriminating.

Perhaps the part of Professor Frith’s interview that I found most refreshing comes towards the end. Despite spending an academic lifetime studying autism, she openly acknowledges the uncertainty that remains within neuroscience and accepts that she herself may ultimately be proved wrong.

That is not weakness. That is science.

Clinical medicine should work in exactly the same way. A diagnosis represents our best explanation based upon the evidence currently available. Sometimes new evidence strengthens it. Sometimes it modifies it. Occasionally we have to abandon it altogether. There should be no embarrassment in doing so.

I frequently remind myself in clinical practice that the most dangerous diagnosis can sometimes be the one we have already made, because once we believe we know the answer we naturally become less inclined to continue asking questions.

None of this diminishes the reality of autism or the substantial challenges faced by autistic people and their families. Nor should it discourage anyone experiencing significant difficulties from seeking appropriate assessment.

But we should resist the temptation to believe that acquiring a diagnostic label is always the final destination. Sometimes it is only the beginning of understanding.

My own approach at The Health Equation — and the Integrated Regulatory Systems Framework that forms the basis of my forthcoming book — is increasingly to ask not simply:

‘What diagnosis does this person have?’

but:

‘What is actually happening to this person?’

What aspects of their developmental history matter?

What physiological systems are contributing?

What is increasing regulatory load?

What is reducing resilience?

What psychological, social and environmental factors are relevant?

What alternative explanations have been considered?

And perhaps most importantly, what can we do that might genuinely improve their function and quality of life?

A diagnosis can be immensely valuable. But medicine works best when the clinician remains curious enough to look beyond it.

Professor Frith’s willingness, after a lifetime devoted to autism research, to reconsider some of her own assumptions is therefore something I find rather admirable. As she essentially concludes in the article below, she may ultimately be proved wrong.

But there we are. That’s science.”

Clinical Disclaimer
This article is intended for general information and education only. It does not constitute medical advice, diagnosis or treatment, and should not be used as a substitute for individual assessment by an appropriately qualified healthcare professional. The discussion reflects the author’s clinical experience and interpretation of the available evidence. Individual circumstances vary, and anyone concerned about their physical or mental health should seek appropriate professional advice.

The autism expert who says people are being misdiagnosed

Uta Frith helped to build the theory of the ‘autistic spectrum’. Now she’s under fire for saying it’s reached the point of being meaningless

Andrew Billen

Uta Frith has spent more than six decades studying autism. After arriving in London in the 1960s and studying psychology at University College London, she became one of the leading figures in autism research and helped develop thinking around what eventually became known as the autistic spectrum. She also played an important role in challenging the damaging historical idea that autism resulted from poor parenting, instead helping establish it as a neurodevelopmental condition with important biological and genetic influences.

Now, however, Frith is questioning whether the concept she helped develop has expanded too far. In a recent paper in Psychological Medicine, she argues that the autism spectrum has become extraordinarily heterogeneous and asks whether this may now contribute to misdiagnosis. Autism diagnoses have risen dramatically over recent decades, while the characteristics encompassed by the diagnosis have broadened substantially.

This position has attracted criticism, including from the National Autistic Society, which argues that increasing diagnosis reflects greater awareness and understanding and warns against creating divisions between autistic people according to their support needs.

Frith nevertheless believes the issue requires scientific debate. She says her concerns have developed gradually and that she has become increasingly convinced that current diagnostic practice has gone too far.

Importantly, she distinguishes overdiagnosis from misdiagnosis. Her argument is not simply that healthy people are being incorrectly told that they have a disorder. Rather, some people seeking an autism diagnosis may have genuine and significant difficulties, but autism may not necessarily provide the correct explanation for those difficulties.

Frith questions why demand for adult and teenage autism assessment has increased so dramatically while the rise in diagnoses among very young children has been less striking. She considers several possible influences, including greater societal awareness and acceptance of neurodiversity, media portrayals of autism, online diagnostic questionnaires and the increasing cultural visibility of autism.

She acknowledges that this is a sensitive area. People may have waited a long time for assessment and may have invested considerable hope in obtaining an explanation for difficulties that have affected them for many years. Questioning the diagnosis can therefore feel deeply personal.

Frith says she has nevertheless heard from a small number of adults who initially felt enormous relief after receiving an autism diagnosis but later began to wonder whether it had actually been correct.

One of her broader concerns is the extraordinary heterogeneity now contained within a single diagnostic category. The children with autism whom she encountered early in her career often had profound difficulties with communication and understanding other people’s mental states. Many contemporary adults diagnosed with autism may be socially and professionally highly capable, presenting a very different clinical picture.

She also reflects on the decision to remove Asperger’s syndrome as a separate diagnostic category and incorporate it within autism spectrum disorder. The intention was partly to simplify diagnostic classification and recognise a continuum of presentation, but Frith wonders whether the resulting spectrum subsequently became too inclusive.

Another difficult issue is the emotional meaning of diagnosis. For someone who has previously been given diagnoses carrying greater stigma, an autism diagnosis may feel like a much more satisfactory explanation of their experiences. It can provide coherence, legitimacy and a new way of understanding their entire life.

Frith does not suggest that clinicians should simply remove diagnoses from people who have found them helpful. Instead, she believes the phenomenon itself deserves greater scientific investigation. Does the relief produced by receiving a diagnosis persist? Does the diagnosis lead to better treatment and support? Most importantly, does it accurately identify the underlying problem?

She is also concerned that popular representations of autism may have influenced public understanding of the condition. Film, television and online culture frequently portray autistic people as unusually gifted, highly intelligent or fascinatingly different. Such representations have helped reduce stigma but may simultaneously have contributed to a simplified cultural picture of autism that differs considerably from the profound disability experienced by some autistic people.

The discussion ultimately returns to neuroscience itself. Despite the remarkable advances made during Frith’s career, she emphasises how little remains definitively understood about the biological mechanisms underlying autism and many other aspects of human cognition.

What is perhaps most striking is her willingness to acknowledge uncertainty. After a lifetime studying autism, she does not claim to possess the final answer. She accepts that future research may produce more refined ways of understanding autism, perhaps identifying meaningful subgroups within what is currently one extremely broad spectrum.

She may also, she readily acknowledges, ultimately be proved wrong.

For Frith, that possibility is not a reason to avoid asking difficult questions. It is precisely why those questions need to be asked.

That, after all, is science.

Andrew Billen, The Times