Atypical antipsychotics in autism: what this Cochrane review found

In one sentence

This Cochrane review suggests risperidone and aripiprazole may reduce short-term irritability in autistic children, while evidence for other outcomes and side effects is still limited or very uncertain.

What the researchers did

The authors reviewed randomized controlled trials that compared atypical antipsychotic medicines with placebo or with another atypical antipsychotic in people diagnosed with autism spectrum disorder. They looked across major databases and trial registries, then combined results using a network meta-analysis for irritability and standard pairwise analyses for other outcomes. In total, they included 17 studies with 1,027 participants: 16 studies in children (996 participants) and one small study in adults (31 participants). The medicines represented in the evidence were risperidone, aripiprazole, lurasidone, and olanzapine.

What they found

• For short-term irritability in children, risperidone and aripiprazole may improve scores versus placebo (mean difference -7.89 and -6.26, respectively), but the certainty was low.
• Lurasidone probably made little to no difference for irritability in the same short-term window (mean difference -1.30; moderate-certainty evidence).
• For important harms and other outcomes, confidence was much weaker: evidence on aggression, weight gain, and extrapyramidal side effects was rated very low certainty in several analyses.
• Signals of benefit were seen for obsessive-compulsive behaviors and inappropriate speech in some short-term comparisons, but these were still low-certainty findings.
• Evidence in adults was very limited, and the review could not support strong conclusions for adult treatment decisions.

What this means for families and therapists

This review suggests that two medicines already used in practice, risperidone and aripiprazole, may help with short-term irritability in some autistic children. At the same time, the evidence on side effects and broader day-to-day outcomes is not strong enough to treat these results as simple “yes/no” answers. For families and clinicians, this supports careful shared decision-making: discuss the target symptom, expected timeline, monitoring plan, and stop rules before starting medication. Medicine decisions should be individualized and reviewed regularly; this summary is not medical advice.

Limitations and what we don't know yet

Most studies were short-term and focused mainly on children, so we still do not know enough about longer-term safety, durability of benefit, or outcomes in adults. Many analyses had low or very low certainty because of bias concerns, small samples, and imprecision. The review also found limited ability to compare subgroups in a clinically useful way (for example, by age bands, support needs, or co-occurring conditions). In addition, not all atypical antipsychotics had enough evidence to judge fairly, and side-effect reporting was not always consistent across trials. So while early evidence points to possible short-term benefit for irritability with specific medicines, stronger and larger trials are still needed to balance potential benefit against risk with more confidence.

---

This is a plain-language summary of Atypical antipsychotics in autism: what this Cochrane review found by Meza N, Franco J.V, Sguassero Y et al., Cochrane Database of Systematic Reviews (2025). Source license: CC-BY-NC-4.0.

It is not medical advice — talk to a qualified clinician before changing therapy.