OCD

OCD and autism occur together in both directions, and at high rates.

Pooling 96 studies of diagnosed conditions in autistic people across all ages, researchers put the share with an obsessive-compulsive disorder at about 9% (95% confidence interval 7 to 10).¹ Restricted to children and adolescents, the pooled figure is a little higher: about 11.6% (95% CI 6.9 to 18.8) of autistic young people meet criteria for OCD.²

Read the other way, about 8% (95% CI 5 to 13) of children and adolescents with OCD have autistic traits or a diagnosis, and about 9.5% (95% CI 6 to 14.7) have a diagnosis of autism.^2,3 The 8% pools questionnaire scores with diagnoses, so it is not the same as a diagnosis rate; the diagnosis-only estimate is the 9.5%.

A Danish national register followed 3.4 million people born between 1955 and 2006. An OCD diagnosis there went with a 13-fold higher rate of a co-occurring autism diagnosis (6.6% against 0.5%).⁴ The same register found that an earlier autism diagnosis predicted a later OCD diagnosis (rate ratio 2.18) and an earlier OCD diagnosis a later autism diagnosis (rate ratio 3.91), and that risk ran in families: parents with OCD were more likely to have an autistic child.⁴ This was one register, so the sequence and the family pattern are not settled.

Clinic samples give higher numbers than population samples, because people who have both are more likely to reach specialist services. In a South London cohort of young people aged 4 to 17, 25% of those with OCD had an autism diagnosis, while 5% of those with autism had an OCD diagnosis.⁵ Among adults, a UK clinic study found that 21 of 73 adult outpatients with OCD (27.8%) met criteria for autism, none of whom had been diagnosed before,⁶ and a specialist OCD clinic in Italy found autism in 149 of 603 adults (24.7%).⁷

A routine is something an autistic person does because it steadies the day or feels right. A compulsion is something a person does to quiet an intrusive, unwanted thought. On paper the two are easily separated. In practice they can look identical, and the same movement can be one thing today and the other tomorrow.

The research shows how hard the line is to draw. A systematic review of 31 studies that compared repetitive behaviors in autistic people, people with OCD and people with both found considerable overlap in their intensity and content; it also found that studies frequently compared only total scores, so there is no established way to tell the two apart with a measure.⁸

Part of the difficulty is that the same behavior can serve a different purpose in an autistic person. In one small study of university students, the compulsions autistic students reported clustered around repetition, counting and hoarding (behaviors that regulate the senses) and were less often the anxiety-driven checking and ordering that marked the non-autistic students. The autistic students' symptoms did not rise with their anxiety as the non-autistic students' did.⁹ This is a single study, in students, and no one has repeated it. It suggests that for some autistic people a compulsion may do the opposite of what it does in classic OCD: it settles the senses directly, where a classic compulsion quiets a particular fear.

A related finding runs the other way. Autistic traits, and especially restricted and repetitive behaviors, are raised in people with OCD, and OCD symptoms track repetitive behaviors more closely than they track social-communication.¹⁰ The authors caution that in adults the raised social-communication scores may partly reflect the effect of long-standing OCD on social life rather than a real difference in that area.¹⁰

Each condition can be missed, and each can be read as the other.

When autism is not noticed, OCD may be the visible problem. In the South London cohort of young people, OCD appeared to be underdiagnosed in autistic youth: 5% of those with autism had an OCD diagnosis, against 25% of those with OCD having autism.⁵ The authors read that gap as OCD going unrecognized next to autism.

When OCD is not noticed, its symptoms may be filed as autism instead. In the same cohort the authors flagged the reverse risk, and a 2020 review of assessment and treatment makes the point directly: there are few validated measures for OCD in autistic people, and the two sets of repetitive behaviors are often hard to tell apart.¹¹ Autistic experience can be read as OCD, or OCD can be attributed to autism and missed.¹¹

The pattern shows up across a person's whole history. In a Dutch register of 1,211 autistic adults, about a quarter (24.6%) reported at least one earlier psychiatric diagnosis, of any kind, that they considered a misdiagnosis.¹² And when adult OCD clinics look for autism, they tend to find it fresh: in the UK clinic study, the autism found in 27.8% of the outpatients had not been diagnosed before. In that study the autistic traits also went with poorer insight into the OCD, which may help explain why the autism went unnoticed.⁶

One finding here is easy to misread. In children and adolescents with OCD, co-occurring autism went with greater impairment on a clinician-rated scale of overall functioning.^2,5 The findings on OCD severity are mixed: one meta-analysis found no difference between the two groups on an OCD scale,³ while a large clinic study found lower self-reported OCD severity in the co-occurring group, which its authors read as a limit of the self-report measure in this group.⁵ Greater impairment, then, does not by itself mean the OCD is more severe.

The standard psychological treatment for OCD is exposure and response prevention, a form of cognitive behavioral therapy. How well it works for autistic people is much less studied than how well it works for everyone else.

The most systematic answer is a Cochrane review. Searching for randomized trials of behavioral or cognitive behavioral therapy for OCD in people with autism up to August 2020, it found just one eligible trial: 46 participants with autism and OCD, comparing CBT against anxiety management alone. That trial pointed toward benefit from CBT on OCD symptoms, but the evidence was of low certainty, and there were too few studies to combine.¹³ A second randomized trial, in 37 children aged 7 to 13 with autism, tested a function-based CBT against treatment as usual on obsessive-compulsive behaviors (the children were not selected for an OCD diagnosis) and found large differences between the groups.¹⁴

Most other evidence is uncontrolled. A study of a CBT program adapted for autistic adolescents with OCD found that about half of its 34 participants responded by the end of treatment, with the gains held at three months.¹⁵ A case-controlled study found that young people with OCD and autism improved with CBT but less than matched young people with OCD alone.¹⁶ Two internet-delivered CBT pilots, one family-based and one therapist-assisted, measured anxiety in autistic youth whose difficulties were anxiety and/or OCD; each improved clinician- and parent-rated anxiety, and neither had a comparison group.^17,18

In children the pattern can run the other way. In a trial cohort of 257 children and adolescents with OCD but no autism diagnosis, those with high autistic traits on a questionnaire did not respond any worse to CBT than those without.¹⁹

For autistic adults, the evidence is thinner still. No randomized trial of CBT or exposure and response prevention for OCD in autistic adults was found as of October 2026. Clinicians who write about the field argue that the trials are missing because the research has not been done, and that this says nothing about whether the treatment works. They point to adapted, therapist-supported programs as a way to widen access.²⁰

Autism and OCD are separate diagnoses, but they overlap in ways that go past behavior.

A systematic review of 50 studies that compared the two found shared heritability, though studies of polygenic risk (many genes each adding a small amount) are still limited. Both are associated with thinner cortex in the temporal lobe, while autism specifically is associated with a thicker frontal cortex. Inflexibility of thinking may be characteristic of both, while how people process faces and hold attention appears to differ.²¹

At the level of brain structure and function, a large meta-analysis of inhibitory-control studies found that both conditions share reduced function and structure in a part of the front of the brain, the rostral and dorsomedial prefrontal cortex, including the anterior cingulate.²² The two also differ: the OCD groups showed increased structure and function in the left basal ganglia and insula relative to the other groups, while the autism groups showed reduced volume in the right basal ganglia and insula relative to the OCD groups.

A third link crosses the two conditions through repetitive thinking. In one study, autistic adults with more insistence on sameness also tended to have more repetitive negative thinking (obsessing and ruminating), which in turn related to more depression and anxiety.²³ The study measured people once, so it shows that these travel together, not which one comes first.

People who have both conditions describe them running together in ways the diagnostic manuals keep apart. Orion Kelly, writing about his own autistic life, puts it plainly: "I use stimming as a way to regulate and soothe myself and in some ways OCD is another form of stimming for me."^c1 Niamh Garvey draws the line differently again: for her, autistic repetition soothes while OCD repetition distresses. She adds that "it is possible to be both autistic and have OCD at the same time."^c2

The hardest question in community writing is the one the research cannot settle: is this a routine or a compulsion? A clinician guide for autistic readers and the people who assess them puts the difference in terms of purpose. In the classic formulation, "a compulsion is always fueled by anxiety," its only job to lower tension, while a stim does not cause distress; and a compulsion cannot easily be swapped for another, while a stim usually can.^c3 Another clinician guide names both errors that show up in clinics: reading autistic experience as OCD, or "attributing OCD to autism and missing it." It quotes an autistic account that captures the stakes: "Autistic routine gives me spoons. OCD takes spoons."^c4

Marian Schembari describes the diagnosis arriving late and by surprise, when a psychiatrist re-read movements she had thought were tics: "What you're describing sounds more like obsessive-compulsive disorder."^c5 And the word "obsession" itself gets used loosely. Alice, in a collection of autistic life accounts, calls her intense interests obsessions ("Obsessions are part of daily life for someone with Asperger's and I still have them despite my age"),^c6 which is how one word comes to mean an intrusive symptom in one context and a loved subject in another. In another collection, a writer objects to that slippage: the same intensity that is called expertise in a non-autistic person, they note, "is labelled as an obsession" in an autistic one.^c7