Summary
Autism, ADHD, dyspraxia, dyslexia, and Tourette’s are classified as separate conditions. They are also diagnosed together so often that the separation starts to look like a filing system rather than a reflection of biology. About 40% of children with dyslexia have another neurodevelopmental condition. The ADHD-autism co-occurrence rate is 40–70%. Dyspraxia appears frequently alongside both. Tourette’s shares basal ganglia involvement with ADHD. At some point, the question shifts from “which conditions does this person have?” to “are these really different conditions?”
This is the overlap problem, and it has practical consequences. A person diagnosed with autism may receive support for social communication but not for inattention. A person diagnosed with ADHD may receive stimulant medication but no sensory accommodations. The diagnostic labels shape what services are available, what insurance covers, and what professionals look for. When the labels fail to capture the actual profile, the support fails too.
What the evidence shows
Genetic clustering
A 2024 study using Genomic Structural Equation Modelling identified five correlated latent genomic factors that cut across diagnostic boundaries: compulsive disorders (including Tourette’s), psychotic disorders, internalising disorders, neurodevelopmental traits (autism + ADHD together), and attention/learning difficulties (ADHD + dyslexia). The study found 49 shared genetic regions and 174 genes between dyslexia and ADHD, 40 of which were previously unidentified.
The Litman-Sauerwald study (2025, Nature Genetics) identified four distinct autism subtypes using genetic and phenotypic data from 5,392 autistic individuals. These subtypes do not map onto clinical diagnostic categories. They cut across the boundaries between autism with and without intellectual disability, and they suggest that what clinicians call “autism” may be several genetically distinct conditions sharing surface features.
Together, these findings indicate that the current diagnostic system carves the neurodevelopmental landscape at the wrong joints. The conditions share more genetic architecture than they don’t, and the boundaries between them may reflect clinical history more than biology.
The transdiagnostic approach
Transdiagnostic research examines shared mechanisms across conditions rather than treating each diagnosis separately. The emerging consensus: neurodevelopmental conditions share significant genetic and neurobiological underpinnings. Attention regulation, executive function, sensory processing, and emotional regulation show atypical patterns across autism, ADHD, dyslexia, and Tourette’s, with more commonality than difference.
A 2024 review argued that the high co-occurrence rates should prompt reconsidering the clinical utility of distinct diagnostic categories. The conditions may be better understood as different expressions of shared underlying neurodevelopmental variation, shaped by genetic weighting, environmental interaction, and developmental timing.
What the labels do and don’t do
Diagnostic categories serve real purposes: they provide access to services, funding, and legal protections. They give people a framework for understanding themselves. They enable research by creating study populations. The autism diagnosis, in particular, has been a powerful organising tool for identity, community, and advocacy (see The diagnosis question).
But categories also constrain. A person with an AuDHD presentation who receives only an autism diagnosis gets autism-shaped support. A person with co-occurring dyspraxia and ADHD who receives only an ADHD diagnosis gets stimulant medication but no motor planning support. The diagnostic system assumes discrete conditions when the biology suggests dimensions and overlaps.
Open questions
Should neurodevelopmental conditions be reclassified around shared mechanisms (attention, sensory processing, executive function) rather than surface-level behavioural criteria? This is a research question with clinical, legal, and identity implications.
How do people with multiple co-occurring conditions experience the diagnostic process? Research on AuDHD women (Craddock, 2024) suggests the experience is one of fragmentation: each condition is identified separately, often years apart, with little integration.
If genetic subtypes don’t map onto diagnostic categories, what should replace the categories? No one has answered this yet.
Key sources
- Genetic clustering across neurodevelopmental conditions (Nature Molecular Psychiatry, 2024)
- Litman-Sauerwald et al. (2025). Four autism subtypes. Nature Genetics.
- PMC review on clinical utility of distinct categories (2024)
- DSM-5 removal of ADHD-autism dual diagnosis prohibition (2013)