It is not ludicrous for mildly and severely impaired to have the same diagnosis
On autism debates and classificatory solutions for social problems
Sometimes I get exasperated by our collective struggle to grasp the basics of how medical and psychiatric diagnoses actually work. More often than not, however, what looks like an argument about classification or diagnosis turns out to be an argument about something else entirely.
The latest example of this comes from the New York Times article “Should the Autism Spectrum Be Split Apart?” by Azeen Ghorayshi (Oct 7, 2025). The parent of a child with autism who is severely impaired and requires round-the-clock care is quoted as saying, “The idea that Elon Musk has the same diagnosis as my daughter is ludicrous.”
This quote was widely shared on social media platforms and, from what I observed, mostly met with enthusiastic approval. Yes, this is a ludicrous state of affairs. Of course, autism should be split apart.
From a diagnostic perspective, however, the statement is obviously wrong. Medical and psychiatric diagnoses encompass a range of severity and impairments within one category. One person with diabetes mellitus may have diabetes so mild that it is fully controlled with diet and exercise, and another person with diabetes may require high doses of insulin, have recurrent episodes of diabetic ketoacidosis, and have a gangrenous foot. Is it ludicrous for both of them to have the same diagnosis?
One person with a viral respiratory infection may be asymptomatic, while another person with the same infection may be on a ventilator in the ICU. One person with mild intellectual disability may function well with relaxed supervision and may even be employed in some capacity, while another person with severe intellectual disability may be fully dependent on others and unable to survive without 24/7 care.
You get the point. What would be ludicrous is to treat a severely impaired person the same as a mildly impaired person. It would be ludicrous to treat a successful businessman with autistic traits the same as an autistic person who is minimally verbal and is intellectually disabled. However, no one is treating them the same, least of all our current classification system.
DSM-5 specifies 3 levels of severity for the autism spectrum: level 1 (”requiring support”), level 2 (”requiring substantial support”), and level 3 (”requiring very substantial support”). Someone in level 3 has severe deficits in verbal and nonverbal social communication skills that cause severe impairments in functioning, and inflexibility of behavior, extreme difficulty coping with change, or other restricted/repetitive behaviors markedly interfere with functioning in all spheres. We can further specify if there is accompanying intellectual disability or language impairment.
What has happened over the past two decades or so is that the public perception of what constitutes autism has been increasingly shaped by those with mild impairment and sometimes those with no impairment at all. The neurodiversity movement has framed the condition as a difference that needs to be accommodated rather than a pathology that needs to be corrected, a conceptual and advocacy shift that seems, whatever its merits otherwise, distant and alien to caregivers of those with severe autism. In addition, parents of children with severe autism, especially those with behavioral aggression, self-harm, intellectual disability, seizures, etc, struggle to access adequate services for their children.
The absence of representation, the mismatch with popular neurodiversity narratives, and the scarcity of needed services and supports are all real problems. None of them are problems of nosology or classification per se. It is not the case that we lack the ability to talk about serious autistic impairments in the current classification. Many of these issues would persist even if the DSM-5 had, for instance, preserved the distinction between autism and Asperger’s. Public attention and social media discourse would still have gravitated toward Aspies, while individuals with “autism” might have remained overlooked and struggling (and their caregivers irritated and annoyed at TikToks of #AspiePride). Simply having a label on its own doesn’t guarantee public concern or attention or representation.
The Lancet commission on autism (2021) proposed “profound autism” as an “administrative term” to apply to
“… children and adults with autism who have, or are likely to have as adults, the following functional needs: requiring 24 h access to an adult who can care for them if concerns arise, being unable to be left completely alone in a residence, and not being able to take care of basic daily adaptive needs. In most cases, these needs will be associated with a substantial intellectual disability (eg, an intelligence quotient below 50), very limited language (eg, limited ability to communicate to a stranger using comprehensible sentences), or both.”
and:
“Profound autism can be associated with complex co-occurring difficulties, including self-injury, aggression, and epilepsy, but is not defined by these factors. Profound autism is not included in the recent revisions of the diagnostic systems (and we are proposing it as an administrative term, not as a formal nosological diagnostic entity)… we offer it for the benefit of autistic people, families, and clinicians and for the purposes of advocacy and description. We hope that its introduction will spur both the clinical and research global communities to prioritise the needs of this vulnerable and underserved group of autistic individuals.” (my emphasis)
The way I see it, the term is similar in function to “serious mental illness,” which is not present in the diagnostic manuals but is a common clinical, colloquial, and administrative term used to refer to conditions where chronic or recurrent psychiatric symptoms significantly impair a person’s functioning (with chronic schizophrenia being the prototypical example).
The “profound autism” proposal is somewhat controversial, and it doesn’t solve the issue that supports and needs still have to be personalized based on the severity of autistic traits, the intellectual and verbal functioning, and the presence of co-occurring medical and behavioral difficulties, given the heterogeneity of profound autism. However, as a practical term, “profound autism” can be helpful in drawing attention to autistic individuals with high clinical needs, despite its limitations and imperfections. If people find the term advantageous and it facilitates advocacy efforts, that’s great. The term will live and die on its own merits; professionals can’t really stop the public from using it, and I don’t believe I am in any position to do so. But it is also true that creating services for autistic individuals with severe impairments or challenging comorbidities does not depend on creating a category of “profound autism.” And similarly, creating a profound autism category will do nothing on its own to improve the state of affairs unless there are investments in expanding services and funding for research.
On the opposite end of “profound autism,” I suspect society may also stand to benefit from an unofficial term that refers to individuals with autistic traits who do not require any clinical or professional support. The term “level zero” seems the most suitable to me for this purpose and provides a natural counterpart to the DSM’s levels.. Level zero already exists as a term online, although it has not been adopted widely. (Seth Underwood, for instance, describes level zero as individuals on the autism spectrum who “require no support whatsoever to operate in normal society” and Monique Moate uses level 0 for subclinical autism.)
What stands out to me in the current debate is the social tension between two groups of stakeholders: caregivers of people with “profound autism” vs “high-functioning” autistic individuals (of the sort active in the neurodiversity movement). Caregivers bristle at the social marginalization of those with profound needs, while many in the neurodiversity movement are frustrated by clinical narratives that default to pathology and misrepresent their lived experience. It is naïve to assume these tensions can be solved by drawing a semi-arbitrary line between “profound” and “non-profound” autism since the real conflict has little to do with classification in the first place. The interests at stake can’t be neatly separated, nor can the parties simply undergo a divorce.
P.S. The bulk of evidence points towards the spectrum nature of autism, with autistic traits existing on a continuum in the population. If it turns out that there are biologically or clinically distinct subtypes that can be meaningfully differentiated, sure, we’ll all have to update how we think about autism, but there is also no guarantee that biological subtypes will align with the interests of the stakeholders.
See also:
