I read a recent editorial on LinkedIn about the way that the diagnostic criteria for neurodiversity require “impairment” in functioning for a person to reach the diagnostic threshold. It was an interesting topic for a paper, and raised a set of broader interlinked questions in my mind about whether diagnosis is a recognition of difference-from-the-norm in the way some people’s brain or body functions, or whether it is about recognising something that is harmful (like a disease), or whether it causes distress or impairment / difficulties in functioning (like a disability or mental health condition). In particular it is interesting to think about how neurodiversity fits into that framework, compared to mental health conditions or physical health conditions. It opens questions about the purpose of diagnosis and whether that is about recognition of difference and understanding ourselves, or for gatekeeping access to certain types of support. It also raises questions about terminology – whether we acknowledge that many patterns of need or difference fall on a spectrum, and what we call a disorder or condition.
As it stands, in the current diagnostic rules (DSM 5 or ICD 11) each diagnosis has a minimum set of symptoms described. For example, in the DSM for a diagnosis of autism the individual must have “persistent deficits in social communication and social interaction across multiple contexts” and two or more areas in which they show “restricted, repetitive patterns of behavior, interests, or activities”. These difficulties must have emerged early in development, and not be “better explained by intellectual disability or global developmental delay”. However, to be diagnosed, the person doesn’t just need to have the neurotype of autism or ADHD but to struggle significantly with work, life, mental health or relationships because of it. Symptoms must cause “clinically significant impairment in social, occupational, or other important areas of current functioning”.
On first glance it seems a pretty clear description that could be consistently applied, but the more I’ve thought about it the more problems I see. First of all, the impairment criteria mean that people could waver in and out of the diagnostic category according to how well they are doing in their lives at that point, how well their environment and network support their needs, or their developmental stage. For example, a person could be autistic as a child, but then find a lifestyle, job, community or other niche in which they are happy and successful, and therefore fall below the diagnostic threshold – until a change or stressful life event means that they become distressed, when they would be diagnosable once again. Or they could have very subtle challenges that would not meet the criteria in primary school, but increasingly struggle during secondary school and meet the criteria for diagnosis, but then escape the sensory overload and/or social demands of that setting and no longer meet the criteria again.
That doesn’t make much sense given that autism, ADHD and other forms of neurodiversity are currently conceptualised as lifelong, inate differences in how people process information and interact with the world, rather than conditions that come and go. Neurodiversity isn’t a disease that temporarily disrupts a person’s normal functioning and can be treated to return them to normal. Neurodiversity is normal for that person; it is intrinsic to the way they experience and interact with the world for the whole of their lifespan. A neurodiverse brain is set up with a much wider focus of attention, or more attention to detail, or more propensity for sensory overload than a neurotypical one. A person doesn’t stop being autistic or having ADHD or a learning disability even if they learn strategies to fit in better with other people’s expectations or cope better cope at school or in the workplace any more than a person who is colourblind stops being colourblind if someone provides glasses that increase the red-green contrast.
For that reason, neurodiversity doesn’t fit into the medical model of diagnosis in the same way as mental health problems that may come and go, where a diagnosis may be helpful during the period when a person is experiencing depression or anxiety or psychosis, but may cease to be relevant at a later point – as for many people the symptoms will entirely resolve (spontaneously or with treatment) and no longer impact them. In the case of a mental health condition it wouldn’t be relevant to continue to use the diagnostic label after their symptoms resolve, as the individual may return to population-normative ways of experiencing the world. If we retained the diagnosis it would lose all meaning, as about half the population would end up with it, but most of those would have no symptoms that differentiate them from those without the diagnosis.
I also wondered how the idea of impairment fits with masking. In some cases a diagnosis could be withheld from a person who subjectively feels like they are struggling enormously if they were able to mask their difficulties enough for them to not impair their ability to function in their workplace or relationships during the assessment period. Or could it be considered a signficant impairment that it is exhausting to mask difficulties? Or that the effort of masking or coping with sensory overload all day at school or work means the person then struggles to function at home, withdrawing or becoming dysregulated in a way that might on first inspection suggest the difficulty is located within the home.
However, despite all the reservations I have raised, it is possible to think about diagnosis as only being meaningful when it applies to functioning, and a label for a degree of symptoms that is outside of the normative range. Many conditions are not binary variables that are either present or absent. Illness is quite often a matter of degree rather than the presence or absence of a condition. For example, a person’s glucose metabolism can indicate that they are becoming insulin resistant, but it is only above a certain threshold that the diagnostic label of diabetes would be applied, and certain treatments would be considered. Likewise our blood pressure or weight can be in the healthy range, or below or above that, but a diagnosis is only given if it is sufficiently high to cause significant health risks. There are numerous variables where we can be tested to see whether we fall into the typical range, or whether further investigation or treatment would be benficial – from our levels of white and red blood cells, iron, vitamin D, cholesterol, hormones, to the functioning of our liver and kidneys, markers of inflammation, infection or tumours. The threshold at which doctors diagnose conditions is typically the point at which symptoms start interfering with functioning and/or the pattern identified is known to cause harm, risk or distress to the individual.
Sometimes medics have to determine appropriate treatment without a diagnostic test being available, or where there is not a direct relationship between symptoms and the scores from blood tests or scans. For example, calcific tendonitis can affect the shoulder joint causing pain, reduced movement or even functional paralysis. However, sometimes people with scans that show a lot of calcification don’t report any symptoms, and sometimes people with very little visible in the scans have a lot of functional impairment. Physiotherapy and/or steroid injections seem equally effective at treating symptoms regardless of what shows in the scans. So treatment is typically determined by impairment alone. Likewise, a woman might have lower levels of oestrogen during or after menopause, and this might be associated with certain symptoms like brain fog, anxiety, hot sweats or disrupted sleep. But there is a degree of variation in what level of oestrogen the woman’s body is used to, and the effect she will experience from it dropping that means some women glide though menopause naturally without much ill-effect, whilst others (myself included) would find it entirely disabling without HRT – and this doesn’t map directly onto the level of oestrogen detected in blood tests. For this reason, it is the pattern of symptoms rather than the blood test results that should guide treatment.
Likewise, depression is something that ocurs along a spectrum from very mild to totally disabling symptoms. As is anxiety. There are even levels of the unusual beliefs or sensory experiences that occur in psychosis that would be considered normative – such as thinking you’ve seen a person who has recently died, hearing an internal voice giving you advice, or having strange or intrusive thoughts that you are able to quickly dismiss. In the same way, arguably we are all neurodiverse to some degree – we each have different patterns of cognitive strengths and weaknesses, different ways of processing sensory information, and subjective experiences of the world. But some of us are able to function in the world more successfully than others, with some people constantly feeling discomfort, distress or having to put in huge amounts of effort to do things that others find effortless. It therefore seems logical that it is the level of impairment rather than the neurotype that determines whether a diagnostic label is appropriate – even though this brings both a level of subjectivity to the judgement of whether the label is appropriate and an acknowlegement that the degree of impairment reflects not just the qualities of the individual, but also the qualities of the environment and network that they function within.
It might be that like height or hair colour, sexuality or personality traits, our strengths and weaknesses, there are ways in which we all vary that are not “conditions” that need to be diagnosed but simply differences between individuals. If we conceptualise neurodiversity as this kind of difference, then recognising it can still be a helpful way to understand ourselves, and how we differ from others. It can also help us find our tribe – the people that share our struggles or our interests and talents.
However, all this theoretical debate misses the elephant in the room of how widening the diagnostic range helps people with more subtle impairment recognise their neurotype, at the cost of negating the experience of those with more profound impairment who had historically been able to convey their needs with a label that now gives most people a false impression of much subtler difficulties. What terminology do parents or carers of people with the “autism” described by Kanner (or at the more impaired end of the spectrum described by Wing and Gould) use to describe their pattern of needs? These individuals may be non-verbal, with significant developmental delay, very atypical ways of interacting with others, and obvious stims/stereotypies that instantly mark them out as different to members of the public. This group includes people with autism who are profoundly disabled to the point that they will need signficant financial and practical support across their lifespan – attending special schools, not being able to communicate their needs except through their behaviour, unable to reach independent living, or have the capacity to make decisions about their lives. Having an appropriate diagnostic label functions as a gatekeeper to accessing this kind of financial and practical support.
The problem being that whilst this group have used the label “autism” to describe their needs for eight decades, this is not the same population as now use the word “autism” to describe themselves on social media, to friends or in the workplace. As the diagnostic threshold has lowered to recognise more subtle impairments the label of autism has been claimed by many quite “high functioning” people to self-describe their struggles in a neurotypical world, and – because this group is more articulate and able to speak for themselves – they have come to dominate the discourse about autism, overshadowing the smaller numbers of people with more profound disabilities who had previously been described by this label. The larger number of people who are now included in the same diagnostic category but were previously given the diagnosis of Asperger Syndrome or High Functioning Autism, or seen as having difficulties that were below the diagnostic threshold, typically don’t have the moderate to profound learning disabilities, or need for support in their daily living skills as those who were diagnosed with Autism a few decades earlier. This population group might be independent, articulate, intellectually able, employed, and/or involved in interpersonal relationships including friendships and romantic/intimate relationships. Whilst it is evident that they do find some elements of life harder than neurotypical people, and it is a very helpful and affirming label that lets many people understand why they have felt different to the norm and have struggled with elements of their life, the level of support they require is not equivalent – despite the diagnosis on paper now being the same. A parallel process of widening the diagnosis has happened with ADHD and other elements of neurodiversity, to the point that 10-15% of people now identify with at least one element of neurodiversity, when historically only 1-2% of people were given these diagnostic labels.
This mismatch has therefore become the grain of truth that underlies some very toxic discussions about the growing rate of diagnosis, and the expanding numbers of people claiming disability benefits and adjustments or support within education and the workplace. With larger, more articulate and more powerful groups of voices advocating for the needs of the able end of the spectrum (who often object to any implication of there being a spectrum or such a thing as high vs low functioning or find such labels pejorative), it is hard for the voices of those representing more disabled individuals to be heard. I can understand that we need to respect the voices of lived experience about what terminology they are comfortable with. But some people don’t have voices in this conversation. And it wouldn’t be honest or fair to pretend that the needs of everyone with autism are equal.
The changing use of the diagnostic label doesn’t necessarily relate to the type or degree of support people require, and therefore it may no longer be able to function as a gatekeeper of services. That might still have to rely on an understanding of impairment – the areas in which we are struggling or require more support than the rest of the population. We can’t support everyone equally, that just isn’t viable. We also can’t allow resource rationing to lead the decisions about terminology or diagnosis. But it will have to be consideration of impairment that helps systems to differentiate what levels of support are appropriate – even if that involves subjective judgements, of functional states that are impacted by the environment, and will need to be regularly reviewed over time. Because without us evolving the right language to accurately reflect levels of need, the most disadvantaged will struggle to get their needs met at all – with tragic consequences for them, their caregivers, and the public purse. And, as Mahatma Gandhi is believed to have said, “the true measure of any society can be found in how it treats its most vulnerable members.”

