For many families and individuals, the path to an autism diagnosis is long and fraught. In 2024, Autistica reported NHS England figures that more than 187,000 people were waiting for an assessment, with some facing delays of up to five years. For those who eventually reach the end of that process, the diagnosis is often experienced not as a burden but as a long‑awaited key to understanding, support, and self‑acceptance. Yet this positive view is not always reflected in public debate.
Parents seeking an assessment for their child may be accused of being ‘pushy’ or making excuses for difficult behaviour. Adults may be told they are ‘not autistic enough’ or that, if they can sustain employment or drive, they do not ‘need’ a diagnosis. In some circles, the growing rate of autism diagnoses has been derided as evidence of a cultural fad or, more harshly, as an ‘unaffordable racket‘.
As a professor of cognitive neuroimaging, and author of The Lost Girls of Autism, I hold a particular interest in this ongoing and contentious debate. Here, I’ll outline its origins; whether diagnosis can ever be ‘too much’, ‘too harmful’ or ‘too easy’; some legitimate questions; and why I think we need to caution against autism becoming a casualty in the ‘over-diagnosis wars’.
The emergence of the overdiagnosis debate
The core definition of overdiagnosis is that it occurs when people are currently symptom-free but are identified as at risk, or on the borderline, of certain conditions. With ever-more sensitive screening tests, and access to predictive genetic information, are doctors handing out too many unnecessary sick notes? The concern is that such pre-diagnoses may confer unnecessary anxiety, create ‘illness identities’ and stretch already limited healthcare resources as the worried well request scans, or assays, or seek uncalled for, and possibly damaging surgical interventions.
When around physical illnesses, such discussions are mainly framed in compassionate tones – sympathetic clinicians worried that they may be doing more harm than good. But when attention turns to the possibility of overdiagnosis in the world of psychological conditions, the tone becomes notably sharper. Even those on the side of compassion display a greater level of cynicism, with subtle (or not so subtle) hints about unearned or pointless diagnoses. There is reference to ‘diagnostic creep’ and the ‘medicalisation’ of normal human variation. Psychological problems are described as just an inability to deal with ordinary distress and anxiety.
Weighing in on the back of this are those with much more trenchant opinions. ‘We’ve fetishised mental illness, so we all want it’, reads one headline. Support for children diagnosed with special education needs has been described as an ‘unaffordable racket’, caused by a system which is ‘patently easy to game’.
Autism is clearly in the crosshairs, targeted because of the dream statistic of a 747 per cent increase in diagnoses between 1998 and 2018. What better evidence is there that there is too much diagnosing going on? This lifelong, neurodevelopmental condition is swept up in the generic overdiagnosis net, tarred with the same brush as allegedly over-diagnosed physical problems.
Is there really ‘too much’ autism?
Is the startling increase in autism diagnoses a reflection that there is an overdiagnosis problem? Have clinicians become over-generous, or careless, or have diagnostic standards slipped?
Those quoting the magic 747 per cent statistic rarely note that, in the 1980s, following a painstaking UK-based investigation of a large range of developmental disorders, there was a deliberate re-calibration of autism’s diagnostic criteria. In the 1940s, Leo Kanner’s seminal description of autism set extraordinarily narrow parameters, capturing only a small group of children with strikingly atypical behaviours. This ‘walled garden’ definition excluded many who nevertheless experienced severe difficulties. By the 1980s, researchers such as Lorna Wing and Judith Gould recognised this limitation and lobbied to expand the criteria to include a broader range of presentations. More recently, in 2013, the DSM‑5 gathered all the earlier subtypes under the single umbrella term of Autism Spectrum Disorder.
Making the definition of autism more inclusive does have a downside. Describing the condition as being on a spectrum was intended to capture the diversity of autistic experience, and the wide range of needs. But, from the outside – perhaps from the perspective of non-expert commentators – the consequence was that the boundaries of the autism concept seemed to be becoming more blurred and imprecise, with no clearly defined boundary between ‘typical’ variations in human behaviours and those that were identifiable as distinctly atypical.
More recently, there has been an emerging awareness that even this more inclusive approach had systematically missed large numbers of marginalised groups – particularly women and girls, and older adults. A survey in 2023 estimated that as many as 750,000 autistic individuals were undiagnosed, suggesting that the total autistic population in England could be over 1.2 million – nearly twice the 700,000 total being reported at the time.
In conjunction with this broadening of the diagnostic criteria and the inclusion of previously overlooked groups, we are seeing a greater public awareness of autism, and how it presents, meaning there is a more sympathetic recognition of the condition and less reluctance to pursue a diagnosis. It should also be acknowledged that, certainly with respect to school-age children, a formal diagnosis is becoming increasingly necessary for autistic pupils to access care plans, consequently increasing the number of diagnoses (and lengthening the waiting lists).
So, yes, there are more diagnoses, but there is no evidence that they are unnecessary, or a sign of slipping standards. They are, conversely, mainly the result of a series of positive changes in the recognition of autism – a correction to a long-standing bias. But here’s the crucial point: many of those new diagnoses aren’t ‘extra’ at all. They are people who were always autistic, but had been overlooked.
Could an autism diagnosis be harmful?
In physical medicine, a diagnosis is an indication that there is indeed a recognisable physical problem, an abnormality which may be associated with pain and suffering, and which could require invasive treatment. With the advent of various screening tests, including genetic assessments, a ‘pre-diagnosis’ is now possible, identifying an individual as currently well but at future risk for an illness of some kind. Altogether, a diagnosis is invariably taken as bad news. It’s understandable that people might be concerned this could cause ongoing anxiety, ill-advised life choices, and the adoption of a negative ‘illness identity’.
But paradoxically, there is clear evidence that, in the world of autism, a diagnosis can often be positively received. For parents who have been seeking explanations for their young child’s difficulties, possibly for months if not years, a diagnosis of autism can offer a longed-for explanation, a gateway to knowledge and understanding, the membership of a community that can offer help and support.
For late diagnosed adults, an autism diagnosis can come after decades of struggle, being ostracised and othered. ‘At last my life makes sense’ or ‘I’ve finally found my tribe’ are frequently reported as responses to an autism diagnosis. Autistic individuals describe their diagnosis as a route map or an instruction manual. A wave of powerful testimonies from late-diagnosed autistic women has been a key driver in the recent awareness of autism’s missing females. Almost universally, they describe their diagnosis as positively life-changing. The subtitle of Emily Katy’s Girl Unmasked: How Uncovering My Autism Saved My Life, could stand as a generic summary of these testimonies.
Yet some commentators are scornfully dismissive of any suggestion of the positive aspects of an autism diagnosis, with Suzanne O’Sullivan writing: ‘A diagnosis may (own emphases) be well-meaning and it may provide a feeling of relief… but there is no proof to say that it does or that the feeling of validation turns into something more meaningful’. Why dismiss a large body of peer-reviewed research and personal testimonies, almost universally celebrating relief and understanding?
Is an autism diagnosis too ‘easy’?
In physical medicine, discussions around diagnostic thresholds tend to centre fairly compassionately on whether too many people are nudged over the threshold from well to ill. With psychological conditions, there’s a scornful sense of ‘diagnostic creep’, and the spectre of the ‘unearned’ diagnosis. In a rather bizarre Catch-22 type scenario, if you have a job and/or a family, can drive a car (yes, literally), can make eye contact – then you don’t ‘deserve’ an autism diagnosis. If you have managed to conceal your lifelong autism-based struggles, then you’re not autistic enough.
Anyone who has been through a full autism assessment (having, perhaps, already waited five years to get there) will attest to the lengthy, complex and rigorous procedures involved. They certainly wouldn’t recognise Nigel Farage’s description of the process: ‘So many of these diagnoses, for Send before 18, for disability register after 18 – so many of these have been conducted on Zoom, with the family GP.’
If anything, evidence suggests that an autism diagnosis is too hard. There is clear evidence of a male bias, especially in the so-called ‘gold standard’ tests for autism. Autism assessment is not a perfect system – the problem of under-diagnosis has already been mentioned – but it is certainly not a system that can easily be ‘gamed’.
What is wrong with ‘medicalising’ autism?
In physical medicine, overdiagnosis has been linked to medicalising or pathologising currently non-existent problems. The same criticism has been employed in the discussions around autism. This echoes the anti‑psychiatry movements of the 1960s and 70s, when the medical model of mental health was attacked for reducing complex struggles to cold biological terms and for legitimising coercive treatments. Yet the clear biological bases of autism aren’t being ‘invented’ by some political agenda or by those thoughtlessly supporting a ‘disease’ model. They are reflecting the reality of the brain differences in this neurodevelopmental condition.
Since the 1970s, research has shown that autism is a highly heritable condition linked to differences in brain development and connectivity. This offers a fruitful explanatory model for the differences in autistic lived experience and why autistic individuals experience the world so differently. Far from being a cause for shame, this knowledge can be empowering. I’ll never forget visiting a residential school for autistic girls and hearing pupils describe with pride their ‘differently wired brains’, pointing excitedly to brain scans that reflected their unique way of seeing the world.
Autistic adults have reported being challenged when seeking an autism assessment, with ‘why do you want a diagnosis, there isn’t a cure’. Yet neuroplasticity – the brain’s ability to change and adapt by forming new connections – is a well-established concept in neuroscience. It may well be a reason behind the effectiveness of early interventions in autistic children. So using the term ‘untreatable’ is also an unnecessary misnomer.
The very geneticists and neuroscientists who have made such understanding possible have been dismissed as harmful biologisers. There are, indeed, concerns within the autism community that reference to autism as a brain-based condition implies a deficit of some kind, rather than difference. But as with all diagnostic processes, it is certainly possible, with care, to alleviate such concerns. And do we really want to go back to the time when the blame for autism was laid at the doors of ‘refrigerator mothers‘?
Questions around systems
There are certainly questions around the current diagnosis of autism. Has the concept of a spectrum done more harm than good? Do we need to go back to autism sub-types? Are ADOS and ADI tests as gold standard as claimed, when there is clear evidence of male bias in their development and application?
The reality is that these questions, and the debate as a whole, are inseparable from discussions about funding and support. Some 30 per cent of individuals with a diagnosis of autism have additional co-occurring intellectual disability; they may be non-verbal and display challenging behaviours, and have very high support needs. There is concern that the explosion of autism diagnostic requests will somehow take away support from this group. This is an extension of the argument in physical medicine that alerting those who are currently well to the possibility that they may succumb to some kind of illness in the future, will clog up medical screening facilities to the detriment of those who are already suffering from such illnesses. With respect to the communities of those with what is now termed ‘profound autism’, they will not be in the queue for assessments, as their difficulties will have been evident from a very early age. The nature of their support needs are not the same as children in mainstream education – both types of needs are, of course, seriously underfunded, but there is no evidence of some kind of ‘robbing Peter to pay Paul’ problems in the system.
The overdiagnoses arguments are also taking place in the context of an upcoming Government review of the funding for support for children with Special Educational Needs, and the legal status of hard-won diagnosis-based care plans. It’s no coincidence that some of the more ‘outraged’ headlines I have shared in this piece are taken from the right-wing press. But removing demand is no substitute for finding the ways to cope with it. This is a long‑overdue recognition of needs that were always there. Those needs should be supported however possible, not dismissed or rationed.
A cause for celebration
The increase in diagnosis of autism should not be seen, in my view, a worrying trend towards ‘a culture of wide-eyed acceptance of mental health gobbledegook’. Rather we can celebrate it as a sign that a civilised society is recognising the struggles that some of their fellow citizens face, and an acknowledgement of their need for recognition and support. The search for the biological underpinnings of autism has rounded out our understanding of the condition and can be part of a sympathetic and evidence-based diagnosis.
Perhaps most importantly, an autism diagnosis is almost universally seen as a gateway to understanding, of positive benefit to self-identity and self-esteem. This is not something that should be rationed or withheld. The over-diagnosis brigade, be they compassionate or outraged, seem to have misunderstood autism and its diagnosis. Autism is not a diagnostic fad but a reflection of human diversity and the world’s need to accommodate it – denying that does far more harm than any so-called overdiagnosis ever could.