BackgroundMedications are widely used among people with autism spectrum disorder (ASD). Few publications, however, document pharmacological prescriptions in autism in France. Our study provides an overview of the use of pharmacological treatments based on parental reports, and assesses the factors that influence their prescription. Patients and methodsWe conducted a retrospective cohort study in France from 2005 to 2007. The parents of 393 children, adolescents and adults with ASD (mean age 12.0 years (SD=8.6), range: 1–45) filled a questionnaire on their child's treatment prescription, including among others the name of the drug, the age of onset and treatment duration, and observed efficacy and side effects. ResultsOverall, 52% (n=203) of individuals have taken medication and 35% (n=139) retained the treatment, with at least two psychotropic medications in 10.5% of cases (n=40). The mean age of the first prescription was 9.4 years (SD=6.4). The mean duration of prescription was 5.0 years (SD=6.7). Atypical antipsychotics (APA) (23%, n=88) and conventional (13%, n=49) were the most commonly prescribed, followed by anticonvulsants (9%, n=36). The number of treatments received was associated with age (U=8908.5, P<0.001), severity of autism (rho=0.29; bootstrap [95% CI: 0.18–0.38]), delay in diagnosis (rho=0.13 [95% CI: 0.22–0.26]), institutionalization (U=7803.5, P<0.01) and treatment duration (rho=0.34 [95% CI: 0.21–0.52]), and was not associated with gender or socioeconomic status. A third of parents (32%, n=127) noted a global efficacy of treatment. Eleven percent (n=45) reported side effects, including sedation (6%, n=24) and weight gain (4%, n=15). The risk of adverse effects increased in case of APA (OR=56.1 [95% CI: 12.4–521.4], P<0.01), polypharmacy (U=2732.5, P<0.01) and off-label prescription (OR=2.4 [95% CI: 1.1–5.1], P=0.031). Each additional prescription increases by 2.4 times the risk of developing side effects. ConclusionThe high prevalence of people with autism who received or pursuing pharmacological treatment questions, given the poor evidence of their effectiveness and their significant side effects.