Parent-defined target symptoms respond to risperidone in RUPP Autism study: Customer approach to clinical trials

Objective: A consumer-oriented efficacy assessment in clinical trials should measure changes in chief complaint and consumer request (symptoms of most concern to patient/caregiver), which may be diluted in change scores of multisymptom scales. Method: In the Research Units on Pediatric Psychopharmacology (RUPP) Autism Network 8-week double-blind trial of risperidone versus placebo, the chief concerns of parents were collected at 0, 4, and 8 weeks (endpoint), in addition to standardized primary measures. Blinded clinical judges rated change from baseline to 4 and 8 weeks on a 9-point scale (1 = normalized, 5 = unchanged, 9 = disastrous); 94 participants had usable data. Results: The most common symptoms identified by parents were tantrums, aggression, and hyperactivity. Interrater reliability was excellent. Mean ratings at endpoint were 2.8 ± 1.2 on risperidone and 4.5 ± 1.3 on placebo (p < .001). Ratings were collinear with Clinical Global Impression-Improvement and Aberrant Behavior Checklist Irritability subscale (primary dimensional measure). Effect size d was 1.4, compared to 1.2 on the Aberrant Behavior Checklist Irritability subscale. Effect sizes varied twofold by symptom category, largest for self-injury (2.11) and tantrums (1.95). Conclusions: Risperidone was superior to placebo in reducing symptoms of most concern to parents of autistic children with irritable behavior. Rating individualized participant-chosen target symptoms seems a reliable, sensitive, efficient, and consumer-friendly way to assess treatment effect and might have clinical application.