To the Editor A randomized clinical trial examined the use of fluoxetine compared with placebo among autistic children and adolescents.1 It represents an important interrogation of common off-label prescribing practices that neglect potential risks. Nonetheless, we are concerned by the use of a potentially flawed primary outcome measure.
The study’s primary outcome measure was the Children’s Yale-Brown Obsessive Compulsive Scale, modified for pervasive developmental disorder (CYBOCS-PDD), a version of an assessment originally designed to measure severity of obsessive-compulsive symptoms.2 The literature shows that autistic repetitive behaviors are fundamentally different from those in persons with obsessive-compulsive disorder. Yet the CYBOCS-PDD treats them comparably, despite the fact that autistic people generally “derive pleasure or relief” from the repetition, unlike the anxiety and difficulty with inhibiting urges that underlie obsessive-compulsive disorder symptoms.3