Mental health

Anxiety and depression are much more common in autistic people than in the general population. Among adults, a meta-analysis of 35 studies found about 27% with an anxiety disorder at the time of study and 42% at some point in life; for depression the figures were 23% and 37%. Its authors note that the studies leaned heavily on clinic samples, and that questionnaires and intellectual disability may shift the estimates.¹

A 2019 meta-analysis of 96 studies across all ages, counting only confirmed diagnoses, put the share with an anxiety disorder at about 20%, a depressive disorder at 11%, obsessive-compulsive disorder at 9%, bipolar disorder at 5% and a schizophrenia-spectrum disorder at 4%; ADHD, at 28%, was the most common.² Studies of clinic patients gave higher figures than studies of whole populations, and the spread between studies was very large.²

A 2019 meta-analysis of 66 studies, also of all ages, estimated that autistic people are about four times as likely as non-autistic people to have depression at some point in their lives.³

National registers avoid selective recruitment, though they count only people diagnosed through health services. In Stockholm, anxiety or related disorders had been diagnosed in 20.1% of 4,049 autistic adults aged 18 to 27, against 8.7% of other young adults; the relative increase was larger in those without intellectual disability. Their non-autistic brothers and sisters also had more anxiety than other young adults, but autistic adults still had about two and a half times the odds of their own siblings, so the link held when family background was shared; in the smaller group with intellectual disability the sibling comparison showed no clear link.⁴ In all of Sweden, by age 25, 77 in 100 autistic women and 62 in 100 autistic men had received at least one psychiatric diagnosis, and 32 in 100 women and 19 in 100 men had been admitted to a psychiatric hospital. Both sexes had much higher rates than non-autistic people of the same sex, and the gap between women and men was larger than among non-autistic people.⁵ Both studies draw on Swedish registers. In adults, a meta-analysis of 53 studies put psychosis at 9.4% and bipolar disorder at 7.5%, higher than the all-age figures above; the studies defined psychosis in different ways.⁶ An updated review of 63 papers, covering about 6.9 million people, found a strong association between autistic and psychotic traits and put the odds of one diagnosed condition given the other at about seven times, but warns that some of this reflects how psychosis questionnaires read to autistic people and how hard clinicians find it to separate the two.⁶⁰ A Dutch cohort shows the measurement problem directly: about a third of 1,686 autistic adults scored above the cutoff on a brief psychosis screen, yet only 5 of the 361 followed up reported a new schizophrenia-spectrum diagnosis, all of them among people who had screened positive, and even so the numbers are tiny; the authors treat the screen as the first step of a two-step assessment, not a diagnosis.⁶² Whole-population data from Manitoba, Canada, agree that the risk is raised: autistic people had about ten times the rate of a recorded schizophrenia diagnosis and about twice the rate of a mood or anxiety disorder.⁶¹

How a study measures mental health changes the answer. In the depression meta-analysis, lifetime depression was 14.4% overall, 28.5% when assessed with a structured interview, and 48.6% when people reported their own symptoms.³ A 2018 review found only one study that had tested how well a depression questionnaire works in autistic adults without intellectual disability, and the evidence it gave was weak.⁷ Since then, a widely used depression questionnaire proved reliable in 947 autistic adults, mostly recruited online, and moderately good at telling depressed from non-depressed people,⁸ and an anxiety scale has been built for autistic adults, whose authors wrote that none had existed before.⁹ Many autistic people find it hard to name their own emotions, which may affect how they answer such questionnaires (see Emotions and empathy).

Some diagnoses look different depending on the tool, and on who does the counting. One meta-analysis found diagnosed post-traumatic stress disorder in about 2.7% of autistic adults over their lifetime, close to general-population figures, and much higher numbers when studies used symptom checklists; its authors suggest the diagnostic criteria may miss how trauma shows in autistic people.¹⁰ A large Swedish register study of people born between 1990 and 2010 measured a different quantity and found a higher relative risk: over the years covered, autistic people were about 4.3 times as likely as matched non-autistic people to receive a new PTSD diagnosis, though the absolute numbers were still small, under 1% of the autistic group. The relative increase was larger in women than in men, and among autistic people the risk was higher again in those who also had ADHD.⁵⁸ A small share of people and a raised relative rate can hold at once; the two findings describe different things.

Personality disorder shows the same split. An early meta-analysis found borderline personality disorder in about 4% of autistic people, within population ranges, and could not test whether one condition is mistaken for the other.¹¹ A UK primary-care cohort found that new personality-disorder diagnoses were about 4.8 times as common in autistic men and 4.6 times as common in autistic women without intellectual disability as in matched non-autistic people; among those with intellectual disability the rate was about eight times higher in women and about twice as high in men, on an imprecise estimate.⁵⁹ Whether this reflects a genuine overlap, a misreading of autistic traits as a personality disorder, or a diagnosis reached when autism was missed is unsettled; the authors call for more work on why the rate rose over the period they studied.

In an online survey of 420 self-selected adults, autistic and possibly autistic people were more likely to report a mental-health diagnosis but less likely to agree with it; a main reason given was that clinicians had confused autistic traits with a mental-health condition.¹² Missed autism is not the whole story of late diagnosis: in Denmark, most people diagnosed autistic as adults (69% of men, 61% of women) had none of the psychiatric diagnoses studied in hospital records before age 18.¹³

Autistic people die by suicide more often than other people: a 2024 meta-analysis of 10 studies, covering about 10.4 million people, put the risk at about 2.85 times that of non-autistic people.¹⁶ Whole-population data from Manitoba, Canada, put autistic people about three times as likely to have attempted suicide, age- and sex-adjusted.⁶¹ Most autistic people who attempted or died by suicide also had another psychiatric diagnosis, and register figures describe diagnosed people only, who may not represent autistic people overall.¹⁵

Thoughts of suicide are common among autistic people without intellectual disability. Pooling 36 studies of autistic and possibly autistic people without intellectual disability, 34.2% had had suicidal thoughts, 21.9% had made plans and 24.3% had made an attempt or other suicidal act.²¹ The pattern holds into midlife and older age: in a UK study of 388 adults aged 40 to 93, the autistic group reported more suicidal thoughts, more thoughts of self-harm and more self-harm than a non-autistic comparison group.⁶⁴

In an online survey of 164 autistic adults, 99 of them women, three things went with higher suicidality scores: self-injury, camouflaging autistic traits, and the number of support needs that were not met.²⁴ In 695 autistic and non-autistic adults, autistic people reported feeling more like a burden to others and less like they belonged, and autistic traits were linked to suicidality through these feelings; a key prediction of the theory behind the study did not hold in the autistic group.²⁵ In 374 autistic adults, feeling trapped linked feelings of defeat to suicidal thoughts, and camouflaging also went with suicidal thoughts.²⁶ An independent Dutch study found that psychiatric diagnoses, loneliness and more autistic traits went with lifetime suicidal thoughts and acts.²³

How much camouflaging adds is unclear. In 180 UK university students, not selected for autism and mostly women, camouflaging did not add to the prediction of lifetime suicidal thoughts once depression and anxiety were counted, though it partly carried the link between autistic traits and suicidality; the authors still treat it as a risk factor.²⁷ A 2024 meta-analysis of camouflaging and anxiety, depression and well-being found every study underpowered to detect small effects and called for studies that follow people over time; it did not pool suicide outcomes.²⁸ See Masking for the full picture.

Standard questionnaires may misread autistic people. Autistic adults understood two items of a common suicide questionnaire differently from how the designers intended.²⁹ An adapted version has been validated for research, not for predicting who will attempt suicide.³⁰

Autistic adults describe mental-health services as hard to reach and poorly suited to them. In an online survey of 200 autistic adults without intellectual disability, most of them women, the main themes were difficulty getting treatment, clinicians who did not understand autism alongside mental-health problems, and a link between poor support and "seeing suicide as their future".³¹ Reviews pooling dozens of studies reach the same view: the most common barrier is that therapists lack knowledge of autism or cannot or will not adapt their approach,³² and services do not adequately support autistic adults "and can even cause additional harm".³³

In an Australian survey with a comparison group, autistic adults reported more barriers to mental-health care than non-autistic adults, and more severe ones. Cost and finding services were the biggest barriers for everyone; barriers on the clinician side (knowledge, attitudes, skills) were especially severe for autistic people.³⁴ In 100 community clinicians in Pennsylvania, intentions and confidence to use cognitive behavioral therapy were lower for autistic clients than for others.³⁵ Among 754 UK autistic adults who did not seek public health care the last time they felt suicidal, common reasons included trying to cope alone, "I did not think they could help me", and long waiting lists. The study had no non-autistic comparison group.³⁶

What autistic adults ask for is consistent across studies: care tailored to the person, sensory changes, remote sessions, longer or more sessions, written or visual information, practical as well as emotional help, and clinicians who know about autism.^37,38

Some talking therapies may help autistic adults, but the evidence is weak and mostly compares therapy with a waiting list or usual care. A 2026 meta-analysis of 35 studies, 1,631 people in all, found a small to moderate benefit for the 16 studies that measured mental health.⁴¹ Most of the interventions pooled reuse an existing cognitive behavioral therapy or mindfulness protocol, and few studies describe their adaptations for autistic people or test whether they helped.⁷² A 2025 network meta-analysis of 67 randomized trials ranked mindfulness-based programs highest for anxiety and cognitive behavioral therapy for depression, at low to moderate certainty.⁶⁹ In English national health records for 8,761 autistic adults who completed a course of psychological therapy, autistic people improved less often (56.1% against 61.7%) and got worse more often (9.2% against 7.2%) than matched adults without an identified autism diagnosis, though most still improved.⁷⁰ One therapy has been tested against usual care in suicidal autistic adults: in a Dutch trial of 123 adults, dialectical behavior therapy reduced suicidal thoughts and suicide attempts by the end of treatment, though the difference was no longer significant at 12 months.⁶⁵ Trauma treatment is the biggest gap: no randomized trial of trauma-focused therapy in autistic adults was found as of October 2026.⁷¹

In England, the number of autistic people without intellectual disability in specialist psychiatric beds rose by about 6 a month between 2015 and 2024, while the total number of inpatients in these beds fell.⁵¹ In interviews with 10 autistic adults about stays on inpatient mental health wards, patients described sensory overload, unpredictability and powerlessness.⁵² In a second study, of 20 autistic adults about an acute mental health unit, the authors concluded that staff seemed to overlook the ward environment as a source of anxiety.⁵³ At one emergency department in Melbourne, 66 visits by autistic adults seen by the mental-health team were more likely than matched visits to involve repeated doses of sedatives, with no difference in physical restraint.⁵⁴ At a Toronto psychiatric emergency department, autism was recorded in only 60.2% of the charts for 1,027 visits by 345 autistic adults.⁵⁵

The Mental Health Act 2025, which covers England and Wales, ends detention for treatment under section 3 of the Mental Health Act 1983 on the basis of autism or learning disability alone.⁷³ The Act received Royal Assent on 18 December 2025, and it defines "psychiatric disorder" as mental disorder other than autism or learning disability, so a section 3 admission needs a psychiatric disorder.⁷³ The change comes into force on a day the Secretary of State appoints by regulations. The only commencement regulations made by October 2026 brought other sections into force, and this change had not been commenced.^73,74 A 2026 consensus statement from clinicians and carers anticipated that detention would move into the less safeguarded Deprivation of Liberty Safeguards framework and that more autistic people and people with a learning disability would be sent to prison.⁷⁵

Police often meet autistic people outside any crime. In an international survey, nearly half of autistic respondents had had police contact; welfare checks were among the most common reasons, and those with contact were more likely to report a mental-health or developmental condition.⁵⁶ In a 2016 survey in England and Wales, autistic adults and parents were largely dissatisfied with police contact.⁵⁷ No study found as of October 2026 has looked specifically at how police handle autistic adults in a mental-health crisis.

Autistic writers often describe their mental-health problems as reactions to a world that does not fit them. Devon Price tells of Daan, whose trauma therapist tried to show him that his fear of people was irrational, when he was still being hurt and rejected almost daily; Price argues that therapy built on battling "irrational beliefs" fits autistic people poorly because many of their fears are realistic.^c1 Sol Smith makes a similar claim about cognitive behavioral therapy and depression.^c3 The research above finds weak evidence for most therapies, adapted or not, rather than a clear difference between them. Some accounts describe help failing before any therapy starts. Vometia, writing in Alex Ratcliffe's collection of autistic adults' accounts, describes mental-health services that said to "just call us", and when Vometia explained that using the phone is hard, said nothing else could be done.^c10

Several writers describe years of other psychiatric diagnoses before anyone suggested autism. Annie, a teacher in Ratcliffe's collection, writes that from the GP to a lead psychiatric consultant, professionals spent 20 years naming a mental illness, from bipolar disorder to borderline personality disorder; Annie was given 13 different medications and spent three weeks on a psychiatric ward before pleading for a referral to an autism specialist, and the diagnosis came within months.^c11 Tracy Turner, diagnosed at 41, writes that through her 20s and 30s she was "always told I had depression, which I always asserted wasn't the true problem", and ties her low mood to years of not fitting in.^c12 Orion Kelly draws the line this way: autism is not a mental illness, but depression and anxiety can occur alongside it, and Kelly has two diagnosed anxiety disorders (see Anxiety).^c13

Writers suggest other routes. Megan Anna Neff writes that "there's no 'one-size-fits-all' therapy for Autistic people", that many respond well to somatic, relational and Internal Family Systems approaches, and that trying to change thoughts directly "can feel invalidating or dismissive of our experiences of marginalization".^c4,c5 Neff and Megan Griffith both recommend finding a neurodivergent-affirming therapist, one who treats autism as part of a person's identity rather than a disorder to fix, and list directories of them.^c6,c7

On suicidal thoughts, Smith writes that some autistic people consider ending things "as the solution to a sort of thought puzzle", a way of feeling they choose to endure. Smith still urges readers to "always seek professional help if you're feeling any impulse toward self-harm", and to "Keep choosing to be here."^c2 Murphy, in Ratcliffe's collection, has had counseling for years for frequent depression and suicidal thoughts tied to being different, and writes: "Counselling has helped me see the ways in which just being me is OK."^c16 Price writes that autistic burnout "presents very much like a major depressive episode", so a therapist may treat the depression and miss the autism (see Autistic burnout).^c1

Writers also question coercive care. Griffith writes: "All mental health treatment should be 100 percent consensual."^c8 Price supports sending mental-health responders instead of police to crises, but adds that mental-health professionals can also "forcibly institutionalize" people of color and disabled people.^c9 Cal Montgomery, writing in Julia Bascom's anthology Loud Hands, describes being institutionalized and then given a choice: consent to the drugs and electroshock that "my team" had already decided on, or face a guardianship case in which their parents would consent for them for the rest of their life. Montgomery writes: "When you choose between traps, you choose the one you might someday escape."^c17

Nonspeaking writers who type or point to communicate also write about mental health. Justice Killebrew, a 13-year-old African American poet, writes that anxious thoughts stayed after he began communicating by typing, and that a therapist helps him with exercises to tame them.^c14 Grant Blasko, who has anxiety and OCD as well as autism and types by pointing with one finger on an iPad, writes: "Please understand that finding a community where you feel valuable, even on your worst days, is critically important for mental health."^c15