Psychiatric medication in autistic adults

Autistic adults are given psychiatric medication more often than adults who are not autistic. In United Kingdom primary care records covering 1997 to 2023, 30% of autistic adults and 14.7% of non-autistic adults had been prescribed an antidepressant in 2023. The records held 172,242 autistic adults and about 34 million non-autistic adults. Antidepressant prescriptions, doses and the length of courses all rose over the period for every group, and prescriptions were highest in autistic adults without an intellectual disability and in women.¹

In the United States, Medicaid claims for 2019 show the same pattern for a different class of drug. Among people aged 14 to 29, 45% of autistic youth with an intellectual disability and 22% of autistic youth without one had been prescribed an antipsychotic, against 19% of youth with an intellectual disability only and 3% of youth with neither autism nor an intellectual disability. Among those taking an antipsychotic, off-label prescribing was most common of all in autistic youth with an intellectual disability.²

Antipsychotic prescribing to autistic people is also high in England, though it fell slightly during the pandemic. A study of the primary care records of 59 million patients found that about 55 in every 1,000 autistic patients were prescribed an antipsychotic in an average month in early 2019, easing to about 51 per 1,000 by late 2021, while rates rose in dementia and care-home groups.³

In the United States, as of October 2026, no medication is approved to treat the core social-communication features of autism. The only approved use of a medicine for autism itself is for irritability, and the medicine is an antipsychotic.⁴ In the US labels, risperidone is "indicated for the treatment of irritability associated with autistic disorder," and its efficacy was "established in 3 short-term trials in children and adolescents (ages 5 to 17 years)." Aripiprazole carries the same indication, "irritability associated with autistic disorder."⁵ Both belong to the antipsychotic group, first developed for psychosis. Neither label describes a trial in autistic adults.

The trials behind the approval were short and in children. A review pooled the randomized trials of antipsychotics in children; for risperidone, across three trials, irritability fell by about 11 points on a standard scale, though the trials varied widely (an I-squared of about 72%). The review called atypical antipsychotics the only drugs with robust, established efficacy for severe irritability in this age group.⁴ A meta-analysis of 11 randomized trials, 1,135 children and adolescents in all, found aripiprazole lowered irritability by about 5 points on the same scale.⁶ In those trials the most common side effects were weight gain, sedation and movement effects, and about as many children stopped aripiprazole as stopped the placebo.⁶

Whether these drugs work in autistic adults is largely untested. A 2026 commentary on irritability in autistic adults describes the treatment used in adults as "largely extrapolated from pediatric antipsychotic trials," and says adults are left underserved by a lack of trials of their own. It argues that irritability is not the same as aggression, that it varies across the lifespan and with support needs, and that better ways to measure it are needed before trials can be run.⁷ No study found as of October 2026 has tested an antipsychotic against a placebo in adults selected for autism, for irritability or for any core feature.

The closest placebo-controlled test, in a group that overlaps with autism but was not selected for it, is discouraging. In a trial of 86 non-psychotic adults with an intellectual disability and aggressive behavior, risperidone and haloperidol did no better than placebo; aggression fell in all three groups, most of all on placebo, and patients taking placebo did no worse at any point. The authors concluded that antipsychotics "should no longer be regarded as an acceptable routine treatment" for that behavior.⁸ The drugs reach autistic adults mainly through the indications approved for children and through off-label use.^7,2

Antidepressants are the class autistic adults are most likely to be given, yet the evidence that they help is thin. A 2026 review of seven randomized trials of selective serotonin reuptake inhibitors (SSRIs) in children, 606 children in all, found the drugs made little or no difference to how the children were doing overall, with a slight rise in side effects; for the symptoms most of the trials targeted, repetitive behavior and anxiety, it rated the evidence too uncertain to judge, and no trial measured depression.⁹ An earlier Cochrane review of nine trials reached the same verdict for children, finding "no evidence of effect" and "emerging evidence of harm," and only limited evidence of benefit in adults, from small studies whose quality it rated unclear.¹⁰

For adults the picture has barely moved. A large randomized trial of sertraline for anxiety in autistic adults, the STRATA trial, has been running; its results had not been published as of October 2026.¹¹ Until it reports, the antidepressant use shown in the records above rests on evidence from children and on small adult studies, not on a trial in autistic adults.

Antipsychotics carry real risks, and the monitoring meant to catch them often does not happen. In a Swedish register of children and adolescents started on an antipsychotic, 399 of 5,441 (7.3%) went on to have a record of a metabolic side effect in the years studied; obesity was the most common. Crude rates were higher in boys than in girls, but the estimated risk relative to risperidone was higher in girls (significant for aripiprazole, olanzapine and quetiapine; in boys only for aripiprazole), with a tendency to higher risk in younger girls.¹² A review of the child trials points the other way on weight, reporting that risperidone carries a greater metabolic burden than aripiprazole; the two sources measure different things, one recorded diagnoses and the other weight.⁴ For autistic people the monitoring gap is measured directly: in one cohort of autistic youth, of 1,538 antipsychotic prescriptions, 847 (55%) were counted as unmonitored against the frequency the guidelines recommend, even though those on an antipsychotic had more lipid and diabetes testing than those who were not.¹³

What the risks look like in autistic adults is less clear, and the two studies disagree. A chart review of 279 autistic adults at one clinic, about a third of them taking an antipsychotic, found no difference from the others in body mass index, triglycerides or total cholesterol.¹⁴ A survey of 545 autistic adults without an intellectual disability found antipsychotic use went with a higher chance of self-reported diabetes, though it could not show cause.¹⁵

Many drugs prescribed for common conditions have anticholinergic effects, meaning they damp down a chemical messenger the brain uses in memory, and autistic adults take a lot of them. In one study of 512 autistic adults with higher support needs, 66% were taking at least one such drug and 31% were taking doses at a level counted as clinically meaningful. Higher potency went with more caregiver-reported memory problems, and with behavior and thinking changes linked to decline, after accounting for age and birth sex.¹⁶ An earlier study from the same group, in middle-aged and older autistic adults, found 48 to 63% taking at least one anticholinergic drug, and higher exposure predicting more reported memory complaints.¹⁷ Both are associations in one group of volunteers, so they show a link, not a cause.

Many autistic people and some clinicians say that autistic people are more sensitive to medication in general and get side effects more often. Some of the evidence behind that picture is real, but specific to particular drugs: reviews find that SSRIs cause slightly more side effects than placebo in autistic children,^9,10 and antipsychotics carry the metabolic risks described above.¹² None of that compares autistic people with non-autistic people on the same drug. What is missing is a controlled study of the general claim. No study found as of October 2026 has compared how autistic and non-autistic people tolerate the same range of medicines. Until one does, "medication sensitivity" is best read as a reason to watch side effects closely and to speak up early, not as a measured finding about autistic people.

Autistic writers treat prescribing as something to go into with your eyes open. Megan Anna Neff, writing for autistic adults, says "Autistic people tend to be more sensitive to medication" and may get side effects more often, and she builds her advice around that: find practitioners who know about autism, keep an open line with them, write questions down ahead of appointments, and "advocate for regular monitoring."^c1 She also notes "there is no prescribed medication specifically for autism," and that the drugs treat co-occurring conditions instead.^c1

A clinician's guide reaches a similar place from the other side. Reviewing the research, Donna Henderson, Sarah Wayland and Jamell White report that autistic participants had lower response rates to some drugs and much higher rates of side effects, and that they have "noticed that their autistic clients seem to respond atypically to many medications."^c2

Not everyone frames the drugs as the answer. In one first-person account, Julie, in her sixties, decides that her depression "may be associated with my Asperger's and so there's nothing to be done anyway," and chooses not to medicate it.^c3