Figure 5. Power calculation for a hypothetical 1-year trial in typical BMD, using NSAA as an outcome measure.

This power calculation assumes selection of BMD patients with a typical BMD phenotype based on functional or genetic criteria, and exclusion of patients with mild/asymptomatic or DMD-like phenotypes; an accepted type I error rate of α = 0.05, and a required statistical power (1−β) = 0.8, and a blinded placebo-controlled design. The number of patients per study arm required for adequate power increases, as the hypothetical difference in 1-year NSAA change between treatment and placebo decreases.