Autism Spectrum Disorder

Autism spectrum disorder (ASD) is a neurodevelopmental condition defined by persistent differences in social communication and social interaction, together with restricted, repetitive patterns of behaviour, interests, or activities, present from early in development and causing clinically significant functional impact.^[1]

No medication treats autism itself. There is no pharmacologic cure for ASD, and no drug reliably improves its core features — social-communication differences and restricted or repetitive behaviours and interests.^[2] The foundation of care is behavioural, developmental, and educational intervention, most commonly applied behaviour analysis (ABA) and related evidence-based programs, started as early as possible and continued for as long as it remains beneficial.^[1] Pharmacotherapy has a real but strictly adjunctive role: treating one specific, clearly identified, functionally impairing symptom that co-occurs with ASD — irritability with aggression or self-injury, overlapping ADHD-type symptoms, anxiety or a rigid/repetitive-behaviour flare, or sleep difficulty — never the diagnosis of autism itself.^[2]

This chapter is organized by target symptom rather than by drug class, because that is how a prescriber and pharmacist should actually approach a patient with ASD being considered for medication: identify the specific symptom causing impairment, confirm behavioural/educational treatment is already in place, rule out a medical or environmental cause of the behaviour, and only then select an agent matched to that one target.

ASD is diagnosed clinically by a qualified specialist — a developmental pediatrician, child and adolescent psychiatrist, or psychologist — using structured history and observation; there is no laboratory or imaging test that confirms it. Diagnostic criteria require persistent deficits in social communication and social interaction across multiple contexts; restricted, repetitive patterns of behaviour, interests, or activities; symptoms present from early in the developmental period, though they may not become fully apparent until social demands exceed capacity; and clinically significant impairment that is not better explained by intellectual disability or global developmental delay alone.^[1]

Severity and presentation vary enormously across the spectrum, from a person requiring very substantial support with minimal or no spoken language, to a person who is independently verbal with subtler social-communication differences. Co-occurring intellectual disability, language impairment, epilepsy, gastrointestinal symptoms, sleep disorders, and psychiatric conditions — ADHD, anxiety, depression, obsessive-compulsive-like presentations — are common and actively shape both the clinical picture and treatment planning.^[2]

• A first-degree relative with ASD — the single strongest known risk factor.
• Older parental age at conception (both maternal and paternal).
• Prematurity or very low birth weight.
• Prenatal exposure to certain medications, most notably valproate.
• A co-occurring genetic condition — for example, fragile X syndrome, tuberous sclerosis complex, or a chromosomal microdeletion/microduplication syndrome.
• Male sex — ASD is diagnosed several times more often in males, though under-recognition in females, who often present with subtler or better-camouflaged social difficulties, is increasingly recognized as a contributor to this gap.
• A sibling already diagnosed with ASD.

Before considering any medication

Confirm the ASD diagnosis was made by a qualified specialist, confirm behavioural and educational intervention is already in place or actively being arranged, and identify the one specific target symptom causing functional impairment — medication is never started to treat "autism" as a whole.^[2]

Rule out a medical or environmental cause of the behaviour first

This step is the single most important, and most often skipped, part of assessment in a nonverbal or minimally-verbal patient — behaviour is frequently the only way an unmet need or a source of pain is communicated.

Baseline assessment before starting risperidone or aripiprazole

Baseline assessment before starting a stimulant, atomoxetine, an SSRI, or melatonin

Follow the Attention-Deficit Hyperactivity Disorder chapter's cardiovascular screening approach in full before a stimulant or atomoxetine. Before an SSRI, screen for a personal or family history of bipolar disorder, since activation and mood switching are real risks. Before melatonin, take a focused sleep history or a brief sleep diary to confirm the difficulty is genuinely sleep-onset or sleep-maintenance in nature, and to distinguish it from a behavioural bedtime-resistance pattern that behavioural strategies alone may resolve.

• Reduce the severity and frequency of the one identified target symptom — irritability/aggression/self-injury, ADHD-overlap symptoms, anxiety/repetitive-behaviour flares, or sleep difficulty — enough to meaningfully improve daily functioning and safety.
• Improve the patient's ability to participate in and benefit from ongoing behavioural and educational intervention, which medication supports but never replaces.
• Minimize adverse effects by using the lowest effective dose, and by reassessing the ongoing need for any psychotropic medication at planned intervals rather than continuing it indefinitely by default.
• Support caregiver and family functioning and quality of life.
• Never expect medication to improve core social-communication differences or restricted/repetitive interests — that is not a realistic or appropriate treatment target for any agent in this chapter.^[2]

• Applied behaviour analysis (ABA) and related evidence-based behavioural/developmental interventions (for example, the Early Start Denver Model in young children) are the foundation of ASD care at every age and for every target symptom — started as early as possible, individualized to the patient, and continued for as long as they remain beneficial.^[1]
• Speech-language therapy and occupational therapy, including sensory-based strategies, support communication and daily functioning and are frequently combined with behavioural intervention.
• Special education services and an individualized education plan support functioning independent of medication status.
• Parent-mediated intervention training teaches caregivers to reinforce communication and behavioural strategies consistently across settings, and is itself an evidence-based intervention rather than only a support for the child's own therapy.
• Structured environment and visual supports — predictable routines, visual schedules, clear transition warnings — reduce the environmental and sensory triggers that often precipitate irritability or anxiety.
• Sleep-hygiene and behavioural sleep strategies (a consistent bedtime routine, a calming pre-bed routine, limiting evening screen time, addressing anxiety or sensory contributors to bedtime resistance) should be tried, and continued, before and alongside melatonin — melatonin supplements rather than replaces this step.
• Cognitive-behavioural therapy, adapted for developmental level and communication style, has evidence for anxiety in verbally fluent patients with ASD and is a reasonable adjunct or alternative to an SSRI trial.
• Combining behavioural and pharmacologic treatment, when medication is genuinely indicated for a specific target symptom, is more effective than either approach alone — present the two as complementary, never as alternatives.^[1]

Pediatric

Almost every agent in this chapter is dosed and monitored differently in children than in adults; formulation-specific pediatric dosing is detailed in the Pharmacologic Treatment Options and Drug Tables sections. Children and adolescents with ASD are frequently more sensitive to both the therapeutic and adverse effects of psychotropic medication than typically-developing peers — start low and titrate more slowly than the standard dose for the same agent used outside ASD.^[2]

Adults with ASD

Irritability, aggression, self-injury, and anxiety can persist into adulthood, and the same target-symptom approach applies. The antipsychotic and SSRI evidence base specific to adults with ASD is smaller than the pediatric evidence base; individualize with specialist input. Because these medications are often started in childhood, actively reassess the ongoing need for continued therapy at each follow-up rather than assuming indefinite continuation is appropriate.

Communication differences — a distinct counselling population

A nonverbal or minimally-verbal patient cannot reliably report a side effect, a symptom of hyperprolactinemia, or a swallowing difficulty with a specific formulation — caregiver-completed standardized behaviour scales (see Monitoring & Follow-Up) and structured caregiver questioning at every visit are not optional extras here, they are the primary way adverse effects and efficacy are actually detected. Administration itself is frequently a genuine barrier independent of "adherence" in the usual sense — taste, texture, swallowing difficulty, and disruption of a fixed routine can all prevent a clinically appropriate medication from actually being taken; see the Clinical Decision Algorithm and Common Errors to Avoid sections for the practical approach.

Intellectual disability

Common alongside ASD and independently increases sensitivity to medication effects, both therapeutic and adverse — apply the same "start low, go slow" principle even more conservatively when significant intellectual disability co-occurs.

Pregnancy and breastfeeding

Reassessing ongoing need

For every agent in this chapter, periodically reassess whether the target symptom, the behavioural context, and the patient's overall functioning still support continued medication — a planned dose reduction or discontinuation trial, done gradually and with the behavioural/educational team's input, is appropriate once the target symptom has been stable for a meaningful period, rather than continuing therapy indefinitely by default.

1. Before concluding any agent has failed: reconfirm the specific target symptom, confirm the dose was genuinely optimized and adherence was adequate, confirm behavioural/educational intervention is still actively in place, and re-screen for an occult medical or environmental cause of the behaviour that would not respond to medication regardless of agent or dose (see Assessment & Diagnosis).
2. Irritability/aggression/self-injury: if risperidone is inadequate or not tolerated at an optimized dose, switch to aripiprazole (or vice versa) and retitrate before concluding the antipsychotic class has failed.
3. If both risperidone and aripiprazole are inadequate, not tolerated, or contraindicated: refer to psychiatry or developmental pediatrics for specialist-level options rather than trialing further antipsychotics empirically in primary care.
4. ADHD-overlap symptoms: follow the Attention-Deficit Hyperactivity Disorder chapter's treatment-failure ladder (switch stimulant class, then trial atomoxetine), adjusted for the lower expected response rate and greater side-effect sensitivity in ASD.
5. Anxiety/repetitive-behaviour flare: if an adequately dosed SSRI trial of sufficient duration (typically 8-12 weeks) does not help, do not simply add a second SSRI — reassess the diagnosis, screen for an undertreated co-occurring condition, and refer to psychiatry rather than continuing to escalate a treatment with an already-weak evidence base.
6. Sleep difficulty: if an adequately titrated melatonin trial does not help, reassess for an untreated contributor (anxiety, pain, obstructive sleep apnea, an inappropriate sleep environment) and refer to a sleep specialist or developmental pediatrics if it remains refractory.
7. Refer to psychiatry, developmental pediatrics, or a specialized ASD clinic for any complex, treatment-refractory, or diagnostically uncertain presentation, and for any patient requiring an agent or combination not covered in this chapter.

• Starting or escalating medication to treat "autism" broadly, rather than one specific, clearly identified target symptom — there is no such thing as an "autism dose" of anything in this chapter.
• Expecting risperidone, aripiprazole, a stimulant, an SSRI, or melatonin to improve core social-communication or restricted/repetitive-interest features — none of them do.^[2]
• Attributing a new or worsening behaviour entirely to autism without first ruling out an occult medical cause — pain, constipation, GERD, dental pain, a seizure, or a sleep disorder — especially in a nonverbal or minimally-verbal patient.
• Starting risperidone or aripiprazole without a documented specific target symptom and the required baseline metabolic and movement assessment.
• Assuming Health Canada has approved risperidone or aripiprazole for this indication — it has not; both are used off-label in Canada, and that should be part of the informed-consent conversation.
• Discontinuing or reducing behavioural/educational intervention once medication is started, rather than continuing both together.
• Using an adult or typically-developing-ADHD starting dose in a small child, or in a patient with ASD generally, without adjusting for the population's greater medication sensitivity.
• Missing prolactin-related, extrapyramidal, or GI symptoms because a minimally-verbal patient did not spontaneously report them — ask the caregiver directly and specifically at every visit rather than waiting for a complaint.
• Treating a genuine SSRI-associated activation or behavioural disinhibition episode as "the underlying anxiety getting worse" rather than as a medication effect requiring dose reassessment.
• Not reassessing the ongoing need for a long-continued antipsychotic, stimulant, SSRI, or melatonin prescription — these are frequently started in childhood and can persist for years without a formal re-evaluation.
• Assuming poor adherence reflects refusal rather than a genuine, fixable administration barrier — taste, texture, swallowing difficulty, or disruption of a fixed routine (see Special Populations and the Clinical Decision Algorithm).

How to use this table: every drug class below is grouped by the specific target symptom it treats, never by "autism treatment" generally — none of these agents improve core social-communication or restricted/repetitive-interest features. Doses require patient-specific confirmation. Select any drug name to open its complete Pharmasist Drug Database page.

Second-generation antipsychotics — irritability, aggression, and self-injury

The only 2 agents with a specific regulatory approval for an ASD-associated target symptom anywhere; both remain off-label in Canada since Health Canada has not issued an autism-specific indication for either.

Stimulant and non-stimulant agents — overlapping ADHD-type symptoms

Full dosing, titration, cardiovascular screening, and monitoring is in the Attention-Deficit Hyperactivity Disorder chapter and applies largely unchanged here; the notes below cover only what genuinely differs because the patient also has ASD.

Selective serotonin reuptake inhibitors — anxiety and rigid/repetitive-behaviour flares

The weakest evidence base of the 4 target-symptom domains in this chapter — reserve for genuinely debilitating anxiety rather than routine use.

Melatonin — sleep-onset and sleep-maintenance difficulty

The best-evidenced target-symptom domain in this chapter for this specific population.

Evidence note: regulatory status, dosing, and comments above reflect current FDA/Health Canada approvals and primary literature, cross-checked against CPS and AACAP guidance.^{[1][2][3][4][6][5]}