Back to List

Title:
Author:

Date:
Source:

The Placebo Responder Rate in Children and Adolescents
Donald W. Lewis, MD; Paul Winner, DO;
Warren Wasiewski MD
Posted: May 2005  
Headache 2005;45:232-239

The key issue facing clinicians conducting clinical trials for the acute treatment of migraine in the pediatric and adolescent patients has been the high placebo responder rate. While adult studies have been able to predictably yield a placebo-response rate about 35%, pediatric trials typically have demonstrated rates to be 50% or higher. Even the most novice student of statistics can appreciate the challenge to designing a study with an anticipated placebo response of greater than 50% to 60%. To demonstrate efficacy with a 50+% placebo-response rate, the differential between active and placebo must be large. Unfortunately, reported studies have typical "active" responder rate has been between 60% to 85%. The burden then falls to the attainment of an "n" sufficient to detect a significant difference. This becomes discouraging, cost-ineffective, and, quite frankly, ponderous.

The purpose of this review is to survey the available literature regarding acute treatment trials, to critically assess the design flaws that have become increasingly apparent and, finally, to comment on "trench" experiences that may be useful in the development of future trials. How can we affect the placebo responder rates?

Conclusions and Recommendations:   There are currently no agents approved by the FDA for the acute treatment of migraine in children or adolescents. Studies must be designed and conducted for the pediatric and adolescent population that can reasonably be predicted to provide statistically significant data to support the widespread clinical use of medicines for the treatment of childhood migraine. The "enemy" has been the high placebo responder rate that has been pervasive in acute treatment trials in children and adolescents.

Final Recommendations:   A combination of these recommendations will need to be incorporated in future studies of the pediatric and adolescent population in order to address the placebo effects. New and innovative study designs must be considered. Multi-treatment, cross over, and "enrichment" designs should be explored. "Enriched" study designs may be particularly useful in the pediatric population. Since the natural history of migraine in the pediatric population is not fully understood, it is imperative to design inclusion criteria with regard to the established migraine history of the patient; e.g., migraine duration of 2 or 4 hours, prior to randomization.