The purpose of this study is to determine if fluvoxamine or sertraline reduce the fequency or severity of aggressive behaviour, obsessive symptoms, or anxiety in young children with autism. The within-patient variability in this patient population using standard neuropsychological instruments will also be determined and a predictor model for SSRI responsitivity based on baseline neuropsychological testing will be developed.

Autism is a neuropsychiatric disorder diagnosed in early childhood. Approximately 10 Canadian children per 10 000 live births suffer from the disorder, which is three to four times more common in males than in females. It is characterized by social and and communicative deficits and restricted, repetitive interests and behaviours. Most autistic children are delayed in the acquisition of both verbal and non-verbal communication skills and many never develop useful language. Three-quarters of autistic children have mild to severe mental retardation and a quarter develop seizures during later childhood or adolescence. Its etiology is heterogenous and there is no cure. Although behaviour therapy is an important tool in management, pharmacotherapy remains a necessity for many children. Current therapy is limited to antipsychotic drugs that can carry an unacceptable risk of chronic neurotoxicity (tardive dyskinesia) or tricyclic antidepressants that have undesireble cardiovascular effects. This study proposes to evaluate the potential benefit of selective seratonin reuptake inhibitors (SSRIs) in autism.

All autistic children whose symptoms are not currently well controlled will be offered entry into the trial. Each child will randomized to eight weeks of fluvooxamine, sertraline, or placebo. If they do not improve on their initial dose, they may have a dose increase and continue in the study for a further eight weeks. Due to the significant amount of within- and between- patient variability, multiple baseline evaluations will be completed prior to the initiation of drug therapy. Parents may choose to continue therapy that was effective for their child; if their child was randomized to placebo, parents may choose to try an SSRI for a period of 8 weeks to assess effectiveness.

Inclusion Criteria:

• diagnosis of autism based on Autism Dignostic Interview -Revised and/or Autism Diagnostic Observation Schedule, depending on which is appropriate for the child's chronological age
• ages 3-10 inclusive
• free of psychoactive medication for at least 3 months prior to entry into the trial

Exclusion Criteria:

• known contra-indications to SSRIs (i.e. hepatic dysfunction)
• Lactose intolerance
• concurrent psychotropic medications (SSRIs can interact with lithium, tricyclic antidepressants, monoamine oxidase inhibitors, and benzodiazepines)
• taking warfarin (SSRIs can increase levels)
• Inability of parents to give informed consent, travel to the clinic visits, administer study medication, or arrange for completion of rating scales by self/school staff